Load Distribution Optimization Of Thermal Power Plant Based On Neural Network Algorithm

As growing energy and environmental problems, energy conservation has become an important assignment for thermal power plant. Based on meeting the regulation requirements, how to optimize the power load distribution of units and to reduce coal consumption for the largest possible has very important significance to reduce the power supply cost and increase the competitiveness of power generation enterprise. In this paper, firstly analyzes the calculation method of unit coal consumption characteristic curve. From the heat balance calculation of thermal power plant, according to boiler efficiency, steam turbine and generator efficiencies, and in consider of the electricity consumption by the power plant itself, calculates fuel consumption under different load of each unit, thus obtains the coal consumption curves. Then, uses the Matlab software and the principle of least square method to fit curves for data from a certain factory. Next, introduces how to make use of Hopfield network to optimize load distribution in details. Making the total load as the objective function and with constraints, obtains load of each unit under the minimum function of the network energy, so the total coal consumption is minimum. Finally, describes how to realize the load optimization of units on the platform of DCS, and introduces how the system configuration and the configuration of software and hardware. Through the research on the load distribution of thermal power plant, obtains the optimized distribution mode that can save coal consumption, which has a guiding significance for saving energy.