39results about How to "Computational Efficiency" patented technology

A system and method are provided for using a computer system to assist in planning a trajectory (960A, 960B) for a surgical insertion into a skull. The method comprises providing the computer system with a three-dimensional representation of the skull and of critical objects located within the skull, wherein the critical objects comprise anatomical features to be avoided during the surgical insertion. The method further comprises providing the computer system with a target location (770, 970) for the insertion within the skull. The method further comprises generating by the computer system a first set comprising a plurality of entry points, each entry point (760) representing a surface location on the skull, and each entry point (760) being associated (2D) with a trajectory (960A, 960B) from the entry point (760) to the target location (770, 970). The method further comprises discarding by the computer system entry points from the first set to form a second, reduced set comprising a plurality of entry points, wherein an entry point (760) is discarded from the first set of entry points if the entry point (760) has an entry angle which fails a condition for being substantially perpendicular to the skull surface. For each entry point (760) in the second set, the computer system assess the entry point (760) against a set of one or more criteria, wherein the set of one or more criteria includes a risk factor based on the separation between the critical objects and the trajectory (960A, 960B) which is associated with said entry point (760)). This risk factor may be calculated by integrating f(x) along the trajectory (960A, 960B) associated (2D) with the entry point (760), where x represents distance along the trajectory (960A, 960B) to a sample point, and f(x) is a function based on distance from the sample point at distance x to a critical object which is nearest to said sample point.

The invention provides an architecture for carrying out multiply-add operation on floating-point numbers or fixed-point numbers. The architecture is characterized by comprising a buffer memory used for carrying out buffer storage on a plurality of input floating-point numbers or fixed-point numbers, the floating-point numbers comprising floating-point indexes and floating-point mantissas; a computing unit array, comprising a plurality of multiplication unit columns at least comprising one multiplication unit, and each multiplication unit column being connected to one addition unit; a floatingpoint control and arithmetic unit module, at least comprising an exponential arithmetic unit, a mantissa shift summator and a shift corrector; and an external input / output interface, used for inputting the floating-point number into the buffer memory or outputting a floating-point multiply-add result. The architecture can be expanded to any precision, and both the calculation efficiency and the calculation precision are considered. Meanwhile, the method is compatible with various memory structures and has extremely high adaptability. The architecture can be changed into any precision fixed-point number operation in the memory through proper modification.

Two-dimensional or three-dimensional images of the distribution of a property of an object are formed by passing rays of radiation through the object and detecting how much each ray is attenuated. The Fourier transform is taken of each individual ray but only the zeroth term of the transform along the path of the ray is retained. Each of these transforms is added into a two or thee-dimensional array. If the three-dimensional distribution is being imaged, the transform is a plane of numbers, which is added into the three-dimensional array at right angles to the path of the ray. The numbers in the array are corrected for the non-uniform density of data. After enough such rays in enough different directions are applied, the distribution of the property is obtained by taking the inverse Fourier transform of the data in the array.

The present invention proposes a short-term load forecasting method for microgrids based on SPSS and RKELM, including steps: (1) collecting online data and periodically updating the historical database (2) preprocessing historical data and extracting load sample characteristics; (3) constructing Offline load forecasting model; (4) Use the Spearman rank correlation method to select historical samples similar to the precursor load of the point to be predicted as online training samples; (5) Calculate the load in the future according to the online training samples and the offline load forecasting model Predictive value. This method uses the rapid simplified kernel function extreme learning machine (RKELM), chaotic particle swarm algorithm and Spearman rank correlation screening (RKELM), and establishes a prediction model including offline parameter optimization and online load; through periodic update of model parameters to Ensure the timeliness of the algorithm, reduce the complexity of online prediction calculations, reduce the amount of historical data storage, reduce computing costs, and can more accurately predict short-term and ultra-short-term loads of microgrids.

The invention provides an azimuth imaging method based on a MIMO rectangular plane array. The method includes a step A of performing backward projection operation on each of four groups of target scattered signals obtained by a receiving antenna when four transmitting antennas are in a time-share transmission condition, obtaining four preliminary radar images and performing modification; a step Bof performing Fourier transform on the modified radar images to a spectral domain, performing truncation and mergence on spectral domain data of the four radar images in the spectral domain and performing Fourier inversion on a comprehensive spectral domain result obtained through mergence and obtaining a final imaging result. The invention effectively solves a problem of the relationship among MIMO array length, array element number and azimuth resolution under the topological structure of the specific sparse rectangular plane MIMO array, and simultaneously gives attention to the operation efficiency. Truncation and mergence are performed on image data in the spectral domain, so that frequency spectrum information of a target is improved and imaging quality is improved. Besides, the computational efficiency is improved and the algorithm real-time performance is ensured by using Fourier transform.

The invention relates to a flight icing numerical value simulation method of a helicopter rotor wing. The flight icing numerical value simulation method is mainly characterized by comprising the following steps that: a vorticity compensation force algorithm is added into a momentum equation and an energy equation in a single fluid model used for vortex movement used for simulating air-super cold water drop two-phase flows under a rotating coordinate system in rotor wing trail traces, and a centrifugal force and geostrophic force algorithm is added into a super cold water drop sliding velocity equation; a water film movement and icing model of the centrifugal force and the geostrophic force is considered; and the algorithms and the models are used for carrying out the flight icing numerical value simulation flow process of the helicopter rotor wing, wherein the model of single fluid two-phase flowing used for simulating the simulating air-super cold water drop movement is realized by a partial differential equation group for describing the movement of air and air-super cold water drop binary mixtures in a rotor wing tail trace flow field, and a two-phase flow slipping speed model. The condition that the quality and energy changes are generated in water films and ice layers because of the rotary movement of the rotor wing is considered in the model of the water film movement and icing process on the rotor wing surface, and the icing shape of the rotor wing surface is obtained.