Developing GA-based hybrid approaches for a real-world mixed-integer scheduling problem

Abstract

Abstract-Many real-worldscheduling problems are suited to a mixecbinteger formulation. The solution of these problems involves t h e determination of integer and continuous variables at each time interval of the scheduling period. The solution procedure requires simultaneous consideration of these two types of variables. In recent years researchers have focused much attention on developing new hybrid approaches using modern heuristic and traditional exact methods. This paper proposes the development of a variety of hybrid approaches that combines heuristics and mathematical programming within a genetic algorithm (CA) framework for a realworld mixed integer scheduling problem, namely the generation scheduling (GS) problem in electrical power systems. The problem is to define onloff decisions and generation levels for each generator in a power system for each scheduling interval. This paper investigates how the optimum or near optimum solution for the GS problem may be quickly identified The results obtained are promising and show that the hybrid approach offers an effective alternative for solving the GS problems within a realistic timeframe.

Evolutionary hybrid approaches for generation scheduling in power systems

“…The composition of two tightly coupled combinatorial and continuous sub-problems of the GS problem represents a typical feature of scheduling problems posed in refinery and resource management [8,11,13,14,18]. Many real-world scheduling problems, such as resource allocation, inventory management, and process scheduling that occur in a broad range of industries are of a mixed-integer nature.…”

An evolutionary generation scheduling in an open electricity market

“…As it can be seen from these figures that the GA solution gives the allocation of generation close to the optimum solution in the most of time periods, except the time periods between 13 and 22, and between 26 and 33. The balancing costs were not affected much by the GA solution in the time periods 26-33, but there is slight increase in the costs for periods [13][14][15][16][17][18][19][20][21][22]. In fact generator 3 significantly increased the price of electricity requiring drastic changes in the generation schedule, and the GA could not finely tune the dispatch variable in the given computational time.…”

“…Previous work by the authors presented GA -based approaches for the generation scheduling problems in centralized electricity industries have shown promising results [2,13,14]. However, the techniques used in the centralized system operation are not ideal for addressing the new market environments.…”

“…This section presents the ISO generation scheduling problem as a mixed integer programming (MIP) problem extending the MIP model of the centralized system presented in [13]. A complete list of the notation used is given below: …”

“…As an attempt to encode both the commitments and dispatch between committed units within a single GA, a problem representation has been designed using a string of binary and real numbers [13]. A sub-string containing one binary variable and one real number has been used to represent a generating unit at a particular time period as shown in Figure 3.…”