Research And Application Of Intelligent Decision Model For Coal-fired Power Plant Purchasing And Dispatching

At present,most decision-making methods of coal-fired power plants are based on the artificial experience of coal-fired power plants,but this method does not take into account the economic and transportation factors.In order to solve this problem,this paper puts forward to the principle of the lowest unit calorific value cost of coal to plant,constructs the optimal decision model of coal purchase and transportation in power plants on the basis of attaching the constraint conditions of mixed coal fuel characteristics,and optimizes the purchase scheme and transportation scheme of coal to ensure the economy and rationality of fuel purchase in coal-fired power plants.The main contents and conclusions of this paper are as follows:Firstly,this paper expounds the technical principle of mixed coal combustion,and two calculation methods for calculating the characteristics of mixed coal,and selects the method with the most application cases and good results at present.Secondly,the coal-fired power station procurement model and coal-fired transportation model are constructed,which are respectively used to calculate the purchase price and transportation price of blended coal.Then,restrain suppliers and transport conditions throughout the procurement process according to the actual situation.Finally,according to the demand for coal quality of coal-fired power plants and the actual consumption of coal in the procurement process,the decision-making model of coal-fired arrival cost(procurement and transportation joint)was constructed.Four algorithms,including genetic algorithm(GA),simulated annealing(SA),particle swarm optimization(PSO)and artificial fish swarm algorithm(AFSA),were selected and their principles and operation processes were introduced in detail.Then,the four algorithms were tested and compared according to the size of the result of the objective function,calculation time,iteration times and error times.From the results,it is concluded that the performance of particle swarm optimization algorithm is best among the four optimization algorithms.Considering that PSO is easy to fall into the trap of local optimization,two ways of improving the particle swarm optimization algorithm were tried.Then,the two improved PSO algorithms and the standard PSO algorithm were tested according to the four performance indexes.The results showed that the simulated annealing algorithm combined with the standard particle swarm optimization algorithm is better than the standard particle swarm optimization algorithm and the linear decreasing weight particle swarm optimization algorithm.Finally,it was decided to apply it to the decision-making system of intelligent purchase and transportation of coal.Complete the construction of intelligent coal purchasing and dispatching system in power station,and describe the information and functions displayed in each interface of the system in detail.After completing the construction of intelligent procurement and dispatching system for coal-fired power plant,an optimized calculation of a coal-fired procurement plan of A Power Plant and B Branch of a certain group was carried out in order to verify the effect of the system.Compare the calculated results with the procurement plans of A Power Plant and B Branch based on manual experience.The results show that the two optimization results are good.A power plant achieved direct purchasing benefits of 13.3209 million,and B branch achieved direct purchasing benefits of 29.3 million.Based on the above results,the intelligent coal purchasing and dispatching system can be successfully applied to coal purchase and dispatch with good effect,which provides reliable quantitative basis for the optimization of the actual coal blending scheme of boiler units,and has important theoretical guiding significance for the optimization of the coal blending scheme of boiler units.