Optimal Scheduling Of Multi-Energy Media For Complex Systems

In the context of excess steel industry,loss and structural reform of supply side,making full use of by-product gas has become an important means for iron and steel enterprises to reduce emissions and increase their profits.In the iron and steel enterprises,the by-product gas system is complex in structure,and the number of equipment involved is large,and the gas consumption characteristics of the equipment are very different.In the process of running,there will be unbalanced problems of production and consumption,resulting in unreasonable utilization of gas resources and the waste of energy.In this paper,a steel enterprise gas system is used as the research object.Based on the modeling of the gas production characteristics of the equipment and the optimization of the mixing ratio of the multi-gas medium,the optimization of the operation timing of the equipment is carried out.The main research results are as follows:(1)The establishment of the gas production,consumption equipment's characteristics of the modelsThis paper analyzes the events and state of gas production equipment and gas consumption equipment.According to the contact between the upstream and downstream production processes of iron and steel enterprises,basing on the production rate of molten iron,the characteristic model of gas production and consumption equipment was established in different state.(2)The establishment of the optimal mixing ratio model of various kinds of gasThis paper studies the optimal mixing ratio of the daily by-product gas of iron and steel enterprises,and puts forward a method of reverse gas decomposition of mixed gas.The volume of the single gas component that contained in the mixed gas that used by each equipment is expressed as functions of heat and calorific value.The optimal mixing ratio model of various kinds of gas uses the mixed gas calorific value and heat of each equipment as the decisionvariables.Minimizing the calorie deviation of each gas consumption equipment is the target,Taking into account the calorific value of the gas equipment requirements,self-provided power plant calorific value,heat requirements and gas volume conservation and other constraints.the optimal mixing ratio model of various kinds of gas contains many decision variables and constraint equations,which makes the traditional multi-objective genetic algorithm is difficult to solve.In this paper,the genetic setting scheme of genetic algorithm is designed,and the range of calorific value of mixed gas is controlled by the range of gene initialization.The results show that the optimal mixing ratio model of various kinds of gas is significantly reduces the number of constraint equations and decision variables.(3)The establishment of the timing optimization model of gas consumption equipmentAt present,the optimal scheduling of gas is less involved in timing(which means the time and duration of the operation of the intermittent production equipment)optimization.Based on the study of characteristic modeling and optimal mixing ratio of multi-gas medium,a 0-1 integer programming model for timing optimization of gas consumption equipment is established.The model minimizes the heat fluctuation and the running state of the consumable equipment.the three kinds of by-product gas volume conservation,self-provided power plant daily input heat constraints,gas equipment and self-owned power plant heat value constraints and flow constraints are taken into account.In this paper,the timing optimization model involves 0-1 integer programming problem(the variable in the model can only be 0 or 1,this kind of integer programming problem is called 0-1 programming problem).In order to reduce the difficulty of solving the model,The genetic setting of multi-objective genetic algorithm is improved.Each set of optimal solutions of the timing optimization model corresponds to a timing matrix(The rows of the matrix represent the time and the columns of the matrix represent the devices.The elements in the matrix are "0" or "1" and the number of "0" or "1" in each row is known,but its position in each row is to be optimized).The traditional solution will make each element of the matrix for a gene.A large number of genes increases the difficulty of solving the model.This paper presents a new way of gene setting.Firstly,It's necessary to determine the number of genes per line.When "1" is less than "0",the number of "1" is the number of genes in the row,and vice versa.Each gene corresponds to an element("1" or "0")in the position of the row.Through the solving of each gene,indirectly,optimized timing matrix can be get.This means can greatly reduce the genetic dimension and reduce the difficulty of solving the model.