| With the " seperation of plant and network,electricity price bidding" pattern formation,the state put forward that gradually transfers the traditional "sub-unit scheduling" model to the "plant scheduling" model,which is plant-level load scheduling model,to improve power plant power efficiency,minimize the cost of electricity generation,thereby enhancing its competitiveness.New energy grid-connected power generation has become the trend of power generation in the future.However,the dynamic and intermittent nature of new energy sources greatly increases the uncertainty of grid operation,especially in the increasing proportion of wind power generation and PV power generation.In the energy storage technology can not meet the demand,the use of controllable power supply to stabilize the fluctuations of new energy is a feasible method.Therefore,it is imperative to study the AGC load distribution technology to adapt to the new energy access.In this paper,it takes the power plant and the new energy as a whole,as a "virtual power plant",the new energy as one of the units.It carries on the study of plant-level AGC load optimal allocation,and the content of this article has three main respects.First of all,this paper put forward that using neural network to forecast the coal consumption because of the fault of way,using coal consumption characteristics curve of unit.The traditional coal consumption forecasting is not suitable for actual production.The least squares method and neural network are used to predict the coal consumption,and the results were compared and analyzed.The results show that the latter is better than the former.Secondly,aiming at the requirements of speed,stability and economy of the AGC load distribution system,this paper designs the corresponding objective optimization function,and puts forward a multi-objective load distribution scheme,which not only satisfies the demand of lifting load,but also obtains greater economic efficiency.At last,the principle of PSO is analyzed in detail,and the algorithm is improved by using simulated annealing.Through MATLAB experiments,an example is given to prove the correctness and feasibility of the load optimization model.The multi-objective load distribution scheme can meet the requirements of the dispatching center and make the power plant obtain more economic benefits. |
