A Comparative Study Of Deterministic Unit Commitment And Probabilistic Unit Commitment

Unit commitment is an important issue in the field of power system operation and development program,and it is significant for the security,reliability,economic operation of power system.Nowadays,the grid dispatching and scheduling departments mainly use the traditional deterministic unit commitment for economic dispatching.The deterministic scheduling method ignores the randomness of wind power.Due to the uncertainty of wind power,when system reserve capacity is insufficient,the phenomenon of "wind power curtailment" will occur.Probabilistic unit commitment,considering the uncertainty of wind power output,hasn't yet formed a complete theoretical system and practical application in engineering.Therefore,it's of great practical significance to research on suitable unit commitment for system with large-scale wind power.Under this circumstance,the paper is focused on the comparative study of deterministic unit commitment and probabilistic unit commitment.The main research contents are shown as follows:1.Based on traditional unit commitment,the paper establishes deterministic unit commitment model.The deterministic unit commitment only considers the wind power prediction value,regardless of the prediction error.The model is transformed into a linear mixed integer programming problem by piecewise linearization.2.The paper establishes the wind power probability distribution model,which is based on the relationship between wind speed and wind power.The model considers the influence of wind prediction value on error.Latin hypercube sampling is used to generate wind power scenarios to describe the uncertainty of wind power.Based on the single wind farm scenarios generation method,"combination-reduction" method is taken to generate the wind power scenarios of the multi-wind farms.3.Based on scenarios method,the paper establishes the probabilistic unit commitment considering the uncertainty of wind power.The model contains two categories of scenarios:wind power prediction scenario and wind power typical scenarios.Wind power typical scenarios are generated by reduction of wind power scenarios set,which have strong characteristic.Based on the Benders decomposition algorithm,the probabilistic unit combinatorial model is decomposed into the unit commitment master model,the feasibility check child model,the scenarios child model and the optimal child model.Through the interaction iterations of the master model and child models,the solution efficiency can be improved.Benders cut is the link between the master model and the child models.4.The paper compares deterministic unit commitment and probabilistic unit commitment from the perspective of economic efficiency of the scheduling strategy,the ability to deal with uncertainty of wind power and wind power acceptance level.Firstly,the generation cost of deterministic unit commitment and probabilistic unit commitment are analyzed qualitatively and quantitatively.In order to compare the ability to cope with uncertainty of wind power,the paper established the check model of unit commitment schemes,and analyzes the two unit commitment scheduling schemes under a large number of simulated wind power scenarios.Based on the two unit commitment scheduling schemes,the wind power acceptance interval model is established to compare the corresponding wind power acceptance interval.Finally,the influence of wind power prediction error on the unit commitment scheduling schemes are compared respectively.The above comparison analysis is of great significance to research more efficient unit commitment considering wind farms.