The Application Of Quasi-steady-state Sensitivity Method In LMP Computation

Locational marginal price(LMP)has been widely used in many electricity markets.When utilizing linear programming models to compute LMP, people need to establish the energy balance equation and the line transmission constraint equation.The key lies in how to compute delivery factor(DF) and generation shift distribution factor(GSDF). At present, people can use traditional methods, such as the incremental net loss method,power trace,MW-Mile,and so on, to solve DF, however, this paper introduces a new formula of DF,which is based on the line loss formulas of imputation theory of dissipation power. It makes some changes to the three conditions of original Shapley value formula, that's symmetry, effectiveness and Additivity.Then it uses nodal active power, reactive power and voltage to represent nodal current in order to acquire new power loss formulas of every line.These new formulas can not only compute DF but also calculate the total system power losses and the power losses assigned to every node. They completely dependent on the real line parameter, have a clear physical concepts and are easy to calculate. At the same time, the delivery factor of the slack reference bus is not 1.So, compared with the existing methods, they have obvious advantages.The traditional methods compute the GSDF with the help of DCPF method, which means the value of GSDF depends on the slack reference bus. It's against the the principle of fairness of the electricity market. In order to overcome the shortcoming above, this paper introduces a new kind of GSDF based on quasi-steady-state sensitivity method. The key lies in how to decide allocation coefficient of every machine to bear the unbalance power of power system. As we know, when there is power disturbance in the power system, the perturbation will be assessed between every generation according to their synchronizing power coefficient.Taking into account the frequency characteristics of power system, this paper adopts synchronizing power coefficient to compute allocation coefficient.When solving the optimization model, this paper still uses the iterative method of AC and DC methods. Firstly, this paper uses Newton-Raphson method to calculate the node voltage magnitude and phase angle, and puts them into the line loss formulas of imputation theory of dissipation power to compute nodal DF of energy equation.Secondly, this paper utilizes the GSDF above to establish the line transmission constraint equation. The whole model is based on the total minimum cost of power generation into a target, takes the last generation optimal scheduling scheme as the basis for the next parameter calculation and continuously update the relevant parameters until the cost difference between the two adjacent iterations satisfys the termination criterion. This method not only retains the accuracy of ACPF method but also keeps the fast speed of DCPF method.At last, taking IEEE30-bus system as example, this paper utilizes linear programming function linprog of Matlab7.0 to simulate every numerical example, and draws up an interface package for LMP Computation based on the GUI function. This package, which has high precision and versatility, is able to compute the power flow based on the Newton-Raphson method and LMPs and its components with and without congestion.