New Mixed Integer Linear Formulation
15 Pages4151 Words209 Views
Added on 2020-04-21
New Mixed Integer Linear Formulation
Added on 2020-04-21
ShareRelated Documents
Name Institutional Affiliation City/ State
Abstract This sheet discuss new mixed-integer linear formulation for the unit commitment problem for thermal units is presented in this paper. Fewer constraints and binary variables needed by the proposed formulation helps in yielding helpful computational savings. The new formulation provides a background for modelling which permits a detailed explanation of time-dependent start-up cost and intertemporal constraints namely
minimum up and down times and ramping limits. The mentioned formulation, mixed-integer linear programming algorithm has been used commercially to solve the unit commitment problem effectively in large-scale cases. NOMENCLATURE Constants A j Factor of the piecewise linear production cost function of j aj, bj, cj Coefficients of the startup cost function of unit j.ccj, bcj,tjcold Coefficients of the startup cost function of unit jCj The unit j shutdown cost.D(k) Demand of load in period kDTj Unit j minimum down timeFj Block slope l of the piecewise linear production cost function of unit jGj This represents the number of period’s unit j that should be primarily online as a result of its minimum up time constraints. Ktj Interval cost of the unit j stair wise startup cost function. Lj Unit j figure of periods that needs to be offline as a result of the minimum down time constraints.NDj Amount of periods of unit j starwise startup cost function.NLj Amount of segments of the piecewise linear production cost function of unit jPj Unit j sizePj Unit j minimum power outputR(k) The spinning reserve needed by period kRDj Unit j ramp-down limitRUj Unit j ramp- up limitSj (O) Amount of period’s unit j stayed offline prior to the period of the time duration.SDj Unit j shutdown ramp limitSUj Unit j startup ramp limitT Number of periods of the time duration.Uoj Amount of time unit j has stayed online aforementioned to the firstly period of time duration.UTj Unit j minimum up timeVj (0) Unit j initial commitment stateVariablescjd (k) Unit j shutdown cost in period k.cjp (k) Unit j production cost in period k
cju Unit j start up cost in period kPj, (k) Unit j power output in period k.I.INTRODUCTION Accurate and efficient tools are needed in the new competitive environment power system to help in supporting decision for resource scheduling. Certain issues such as determining when to shut down or shut up generating units and dispatching online generators to attain the demand of the system has been solved traditionally in thermal unit commitment problem. The problems associated with scheduling of generation in the recent market are solved by independent system operator (ISO) and these problems are identical to those of unit commitment in unified non-competitive. There is relevancy in the competitive power industry when itcomes to solving traditional and centralized unit commitment problems. The potential savings in operating cost has led to active research on mixed- integer, large scale, non-linear programming and combinatorial problem. Many techniques for example dynamic programming, heuristics, mixed-integer linear programming and Langrangian relaxation simulated annealing were enacted to solve the issue of savings in operating cost. As a result of solving large-scale problems, lagragrian is the most widely used technique among the mentioned techniques. The solving of unit commitment problem was the first action to be carried out by the Mixed-integer linear programming (MILP and its assurances merging to be the best solution involving a few number of steps and giving out modelling framework which is accurate and flexible. Developing a branch-and-cut algorithm which is an example of an effective example of mixed integer linear software has enhanced commercial solvers with large-scale proficiencies. Its formation was grounded on binary variables which are three sets correspondingly, the shutdown, and startup and on/off states for every time period and every unit. The paper's objective is to provide a substitute mixed- integer formulation for the thermal unit commitment problem represented as MILP-UC. This model is also used in setting up problems incurred in place dealing with electricity such as market-clearing procedure which is solved by ISO. Thus, MILP-UC is beneficial to the market agents. This paper aims at solving realistic application through numerical experienceand formulation of MILP_UC demanding less binary variables and constraints to help in decreasing the computational problem of current MILP approaches. Models lacking quadratic features are referred to as mixed integer linear programming problems and they are solved as a result of algorithm description. Mixed- integer linear program is a problem having;Bounds and linear constraints with exclusion of nonlinear constraints.Limiting some components to have integer values.Objective function in linear Intlinprog Algorithm It constitutes of Algorithm overviewLinear Program Pre-processingBranch and boundsCut GenerationLinear ProgrammingMixed- integer Program pre-processingHeuristics for Finding Feasible Solutions
End of preview
Want to access all the pages? Upload your documents or become a member.