56results about How to "Guaranteed global optimality" patented technology

The invention discloses an optimal configuration method of a distributed combined cooling heating and power system. The optimal configuration method of the distributed combined cooling heating and power system includes the following steps of establishing a digital model library of various energy utilization and conversion forms of the distributed combined cooling heating and power system, wherein the model library comprises a energy model, a cost model and a pollutant discharge model; establishing a feasible configuration scheme library according to load requirements, constraint conditions and combination screening strategy, setting total configuration schemes in the configuration scheme library as N; conducting annual hourly operation strategy optimizing to each configuration scheme in the configuration scheme library according to an annual load demand curve, until working out annual operation cost, annual once energy consumption and annual pollutant discharged quantity of an ith configuration scheme, wherein the i is greater than or equal to N; and selecting a scheme which is the minimum in annual operation cost, the minimum in energy consumption and the minimum in pollutant discharged quantity or a scheme which is arbitrarily combined the minimum annual operation cost, the minimum energy consumption and the minimum pollutant discharged quantity as the best configuration scheme.

The invention discloses a method for predicting the remaining life of a rotating machine based on an integrated GMDH framework. The method includes the following steps: S1, collecting multiple sensor data of a plurality of rotating machines of the same type from normal operation to failure, and passing the data processing to obtain the training data set W; S2, divide the data set into three different GMDH prediction networks respectively; S3, use the prediction output of the three GMDH networks on the training samples as a three-layer BP The input of the neural network trains the BP neural network, and the BP neural network is used to integrate the prediction results of the three GMDH networks; S4, using the integrated GMDH framework to predict the remaining life of the rotating machinery, calculate and output the remaining life prediction value. Compared with a classic LSTM network and a single GMDH network, the present invention can effectively improve prediction accuracy and generalization ability, and has greater practical guiding significance.

The invention relates to a route planning method for an autonomous underwater vehicle aiming at undetermined mission time. The route planning method is characterized by comprising the following steps of: encoding a route planning problem to obtain a population with a scale of N; estimating population fitness and selecting N good individuals from the current population; carrying out interlace operation on the good individuals at a probability of pc; carrying out mutation operation on the interlaced good individuals at a probability of pm; carrying out an elicitation type mutation operation to obtain an optimal individual; and decoding the optimal individual according to an encoding rule in the step 1, so as to obtain a planned route of each AUV (Autonomous Underwater Vehicle). The route planning method for the autonomous underwater vehicle aiming at the undetermined mission time provided by the invention can be used for planning the routes of a plurality of AUVs, so as to realize cooperative route plane of a multi-AUV system, and ensure the global optimum of a planning result. Furthermore, the route planning method has the characteristics that use is convenient and generality is strong, and can be applied to the route plane of a robot and the like.