| Under the environment of Electric Power Market, the advantage and disadvantage of bidding model has directly effect the fairness, justice and reasonableness of Electric Power Market' operation process. It also has very important theoretical value and application prospect that how to using better and more universal bidding algorithm which can effectively reduce the cost of electric energy in order to reduce the price of electric power.Block bidding model is a new bidding mechanism aiming at the disadvantage of hourly bidding model. The Merit-Order Method which been commonly applied at Block Bidding Power Market can only process step-like bidding curves, but be useless in linear bidding curves. This paper presents a block bidding algorithm base on genetic algorithmic, which provides a new way to solve this problem. According to the characteristic of block bidding mechanism, the coding and genetic operators are designed. This algorithm also adopt self-adaptively crossover rate and mutation rate, niche technology to keep population's diversity, avoid premature and enhance local search ability. Basing on the result of the algorithm, the daily generation scheduling is planned in which a whole day time be divided into 48 points. During this process, the load prediction curve is partitioned according to ramping constraints.On Matlab software platform, the improved genetic algorithm has been applied to the cases using linear and step-like bidding curves. The results prove that optimizing effect of the algorithm is slightly difference according to the difference of bidding curves' properties, but under the both two situation gains optimized power purchase cost. The results prove the effectiveness and feasibility of this algorithm, and has no limits of bidding curves' form. |
