40results about How to "Solve the "curse of dimensionality" problem" patented technology

The present invention relates to a distributed safety characteristic selection method based on power grid operation data, and belongs to the power grid safety operation technology field. The method comprises the steps of establishing a power grid power flow characteristic set INP according to the power grid operation data; for the established power grid power flow characteristic set INP, obtaining the correlation among the power grid power flow characteristics, and then obtaining the correlation degrees among the power grid power flow characteristic subsets, and merging the power grid power flow characteristic subset having the maximum correlation degree to finish the correlation grouping based on the power grid power flow characteristics; at different calculation nodes, carrying out the conventional characteristic selection on each power grid power flow characteristic subset INPi in a distributed manner, and integrating the calculation node results at coordination nodes to finish the distributed safety characteristic selection. According to the present invention, the dimensions of the power grid power flow characteristic set are reduced substantially, so that the curse of dimensionality problem in a conventional characteristic selection algorithm is solved to a large extent, and the calculation efficiency is improved.

The invention relates to a word computing method and a word computing device based on a broad general type-2 fuzzy set. The method comprises the steps of building a broad general type-2 fuzzy set word model; building a univariate rule module, measuring proximity between an actual input assessment criteria linguistic value and a corresponding univariate rule module rule first component to serve as a weight and a rule consequent, and aggregating to acquire an intermediate output word linguistic value; operating to acquire a univariate rule module weight via a constrained multi-objective optimization algorithm, and aggregating the univariate rule module weight with the intermediate output word linguistic value to acquire a final output word linguistic value; determining a plurality of object relations to be evaluated through sorting via a fractional function method, or computing proximity between the final output word linguistic value and the known output word linguistic values and selecting the value with the maximum proximity to serve as the actual output word linguistic value of the object to be evaluated. According to the method and the device provided by the invention, the problem of curse of dimensionality caused by rapid increase of fuzzy rules is solved, word computing efficiency and accuracy in artificial intelligence equipment are effectively improved, and wider application prospects are provided.

The invention discloses a four-rotor unmanned aerial vehicle formation reconstruction control system based on ADP and obstacle avoidance mechanism, including: a limited-time comprehensive disturbance observer based on sliding mode, which estimates the external comprehensive disturbance, and simultaneously reflects it according to the disturbance estimation value design The cost function of uncertainty, tracking error and control quantity; the formation reconstruction controller based on ADP uses the neural network to approximate the optimal cost function and solve the optimal control law, and uses the potential energy function to complete the design of UAV collision avoidance items ; The finite-time attitude tracking controller based on the sliding mode performs stable response tracking control on the yaw angle, roll angle and pitch angle of the UAV attitude, and realizes the formation maintenance and formation transformation of the multi-UAV formation in the interference environment control. The present invention adopts the above-mentioned four-rotor UAV formation reconfiguration control system, realizes multi-UAV formation generation, maintenance and reconfiguration control under the influence of comprehensive disturbance, and ensures high-quality flight of UAVs.

The invention discloses a ligand molecule massive characteristic screening method in drug design. In the drug molecule virtual screening based on the ligand, dimensionality (every dimensional characteristic represents a substructure) of the ligand molecule fingerprint feature generated by using the most popular ECFP method at present is massive and even ten thousands of dimensions due to the large number of ligand molecule, so the method can meet the problem of 'dimensionality disaster' in actual task. The method screens the massive ECFP molecule fingerprint features by using the LASSO method based on EDPP rule, and acquires related characteristics of the ligand molecule by using a robustness selecting method. The activity of the ligand molecule is often related to a few number of substructure; the method can rapidly and largely remove uncorrelated characteristics, select related characteristics of robustness, solve the problem of the 'dimensionality disaster', acquire the substructure related to the ligand activity and push the wider application of the ECFP method in the drug design.

The invention discloses a task constraint-based spacecraft attitude control system on-orbit reconstruction method and belongs to the technical field of reconstruction of a spacecraft attitude controlsystem. Task constraint conditions are introduced into design of a utility function and a performance index function, iteration control rules of an HJB (Hamilton-Jacobian-Bellman) equation are determined on the basis of the utility function and the performance index function, and a neural network is trained in an ADP (Adaptive Dynamic Programming) framework to conduct fitting of a target functionunder the iteration control rules and then to determine an optimal reconstruction strategy. In the case of faults, the optimal reconstruction strategy ensuring that set tasks can be carried out as successfully as possible is adopted, and task completion capabilities and fault handling capabilities of a spacecraft are improved.

The invention, which belongs to the technical field of artificial intelligence and control, relates to a neural network and enhanced learning algorithm, particularly to an inverted pendulum control method based on a neural network and reinforced learning, thereby carrying out self studying to complete control on an inverted pendulum. The method is characterized in that: step one, obtaining inverted pendulum system model information; step two, obtaining state information of an inverted pendulum and initializing a neural network; step three, carrying out and completing ELM training by using a straining sample SAM; step four, controlling the inverted pendulum by using an enhanced learning controller; step five, updating the training sample and a BP neural network; and step six, checking whether a control result meets a learning termination condition; if not, returning to the step two to carry out circulation continuously; and if so, finishing the algorithm. According to the invention, a problem of easy occurrence of a curse of dimensionality in continuous state space as well as a control problem of a non-linear system having a continuous state can be solved effectively; and the updating speed becomes fast.

The invention discloses a method for optimizing a minimum broken point set based on characteristics of spatial distribution. The method for optimizing minimum broken point set based on the characteristics of the spatial distribution comprises the first step of dividing the whole network into k communicated subnets at the positions of a node-cutting point and an edge-cutting set, the second step of dividing four gathered subspaces of the i connected subnets into a first subpopulation, a second subpopulation, a third subpopulation and a fourth subpopulation, the third subpopulation and the first subpopulation are mutually symmetrical, the fourth subpopulation and the second subpopulation are mutually symmetrical, the third step of working out minimum broken point sets of the first subpopulation and the second subpopulation by means of binary particle swarm optimization, the fourth step of obtaining minimum broken point sets of the third subpopulation and the fourth subpopulation according to the spatial symmetrical characteristics of the minimum broken point sets, the fifth step of repeating the step 2 to the step 4 and getting the minimum broken point sets of the k connected subnets, the sixth step of conducting union set operation on the minimum broken point sets of the k connected subnets and obtaining multiple sets of the minimum broken point sets of the whole net. The method for optimizing the minimum broken point set based on the characteristics of the spatial distribution solves the problem of 'curse of dimensionality', and rapidly obtains multiple sets of the MBPS which are quite different in positions by means of the spatial symmetrical characteristics.