The Research On Spare Parts Consumption Forecasting Of Thermal Control System In The Electric Power Plant

The consumption of spare parts which are fixed in the thermal control system in the electric power plant is highly uncertain, therefore, in the field of spare parts management, the plan of spare parts demand in most of the enterprises is always a reporting requirements planning based on years of production experience. In order to guarantee the normal operation of equipment, reporting requirements planning is often enlarged, besides, when the material supply and departments starts to purchasing materials , according to the judgment between the plan submitted to them and their own experience, the number of purchases may lead to further amplification, then resulting in a large number of spare parts inventories, slower inventory turnover, a great deal of inventory cost, all of these results can directly impact on business economic benefits.Only by accurately forecasting the trend of consumption of spare parts, the spare parts reserves can be reasonably determined, and then lower inventory costs. In this paper, for characteristics of the spare parts of thermal power plant control system and shortcomings of the status quo in the management, the consumption of spare parts of the thermal power plant control system has been researched. By the research on inventory management of spare parts of current thermal power plant control system, and in conjunction with the status quo about the research on spare parts consumption forecasting method at home and abroad, this paper puts forward a innovative ABC classification model of spare parts based on BP fuzzy neural network, and various types of spare parts consumption models based on the innovative ABC classification. The innovative ABC classification model of spare parts based on BP fuzzy neural network, takes into account the three important aspects of spare parts, such as the key indicators,the consumption indicators and the economic indicators. The design of the model is a four-tier structure of fuzzy neural network: There are three input layer nodes, corresponding with the score of the key indicators,the consumption indicators and the economic indicators; Nine subjected layer nodes, corresponding with the subjected degree which the key indicators,the consumption indicators and the economic indicators belong to the three categories of spare parts; Three output layer nodes, namely A, B, C; the choice of hidden node number is in accordance with the methods proposed by U.S. scholar Hebb, the hidden node number of this article is five. Through training the network, the establishment of the ABC classification model of spare parts based on BP fuzzy neural network is in line with the actual needs, and then based on this model; this article put forward various types of spare parts consumption models. Research shows that the method which is put forward by this article is a superior to traditional forecasting method, such as by virtue of the experience of forecasters. It effectively reduces the factitious error brought by inexperienced staff and improves the accuracy and usefulness of the forecasting results.To sum up, according to features of the spare parts of thermal power plant control system, this article puts forward a innovated ABC classification model of spare parts based on BP fuzzy neural network, and various types of spare parts consumption models based on the ABC classification. Finally, summing up the full text, and looking up the prospect of the spare parts consumption forecasting.