Research of self-power Generation Scheduling Model Base on Multi-objective in Iron and Steel Enterprises

With the targets of the minimum cost of power generation and the lowest rate of gas emission in iron and steel enterprises, a multi-objective self-power generation optimal scheduling model was built based on multi-objective particle swarm optimization the research of coupling relationship of gas and power. And by using the hierarchical decomposition method, the model was broken down into two parts: optimization of gas system and optimization of thermal and power system. The case analysis indicated that: the model could distribute the energy of gas and power reasonably, safely and efficiently when the production condition was changed, and improve the energy utilization efficiency.

generation, diffusion and others. Therefore, a multi-objective self-power generation optimal scheduling model is proposed in this paper which based on multi-objective particle swarm optimization (MOPSO) the research of coupling relationship of gas and power.
According to the features of gas pipeline network system, combined with the actual scheduling situation, the by-product gas system is modeled and analyzed based on MOPSO rules and priorities. The results proved that this model is useful to reduce gas emission, achieve the balance of the entire gas system and have important theoretical and practical benefits.

Mathematical description of PSO [6]
PSO is a population-based optimization approach. The basic idea behind the algorithm is to use a collection of particles to explore the fitness landscape of a particular problem. Each particle is a vector that describes a candidate solution, and can be evaluated along several quality dimensions. The algorithm is iterative, and at each iteration each particle moves through the fitness landscape according to its current fitness values as well as those of nearby particles, and the swarm as a whole.
Formal description of multi-objective optimization problem is shown as below: Where X is the decision vector, Y is the target vector, ( ) f X is the objective function, ( ) The exact steps of the PSO algorithm for the single-objective case are as below: (1) Initialize the swarm; (2) For each particle in the swarm: (4) Repeat.

Modelling
Based on the above algorithm, with the target of the minimum cost of power generation and the lowest rate of gas emission in iron and steel enterprises, a multi-objective self-power generation optimal scheduling model was built based on MOPSO.

Problem description
The byproduct gas generated from iron making and steel making are supplied to gas users after pressurization. Surplus gas is sent to power plants' boilers for power generation. The small quantity of gas imbalance frequently aroused by fluctuation of working condition is absorbed by gas holder, when surplus gas is insufficient, the boiler will be switched into the model of coal firing or gas coal co-firing. Detail as Figure 2.
Generally, heating furnace and boiler have fuel interchangeability, that's to say, if allowed technically, such devices may use different fuels; when different fuels are used, the devices have differences in fuel consumption of unit output.
Therefore, reasonable gas scheduling may lower the energy consumption of the system. Meanwhile, similar devices running in parallel have differences in conversion efficiency when manufacturing different energy resources, so self-power generation scheduling model optimization may be used to solve the following problems: (1) In the precondition of knowing enterprises' production plan and equipment maintenance plan, we may reasonably allocate various fuels (blast furnace gas, converter gas, mixed gas, coal) to different energy consumption devices; in condition of meeting production demand, we shall minimize the loss of energies caused by the mixture of energy resources of different qualities, lower the energy diffusion of the system, and control the energy consumption of the whole enterprises' fuel system within the minimum scope;

Objective function
The objective function of steel and iron enterprises' multi-energy cooperated optimization model is as shown in