30 results about "Linear correlation analysis" patented technology

Embodiments of the present invention provide a method and device for diagnosing hidden faults of a 4G base station. The method includes: acquiring relevant dimension data, integrating the relevant dimension data to obtain a data index set; and obtaining the data index based on a linear correlation analysis method According to the correlation coefficient between the index items of the set, reduce the dimension of the data index set according to the correlation coefficient to obtain a dimension-reduced data set; input the dimension-reduced data set into the trained fault location model to obtain the fault identification result. The embodiment of the present invention provides a method and device for diagnosing hidden faults of 4G base stations, which introduces AI machine learning technology, and forms an intelligent hidden fault troubleshooting model by mining the correlation between various deep-level factors of hidden faults of 4G base stations. , to realize the automatic identification and diagnosis of problems.

The invention discloses an efficient control method for realizing optical network construction on the basis of linear correlation analysis, and belongs to the technical field of communications. According to the method, multi-dimensional optimized decomposition is carried out on optical access network hotspots by adopting linear correlation of feature vectors, and the most suitable construction models are matched to satisfy the construction demands of the hotspots. According to the method, linear matching of the optical access network hotspots and the construction optimized models is newly provided, an approximation set is found, a Bayesian network is constructed to carry out decision judgement, and the optimum construction models are selected to realize the efficient construction of the hotspots.

The invention provides a method for quickly identifying the layered information of sediment pollution in shallow lakes by shallow profile. First, the shallow profile device scans and collects columnar samples, then performs columnar sample analysis and linear correlation analysis, and finally draws a contour map to obtain Sediment contamination stratification results. The present invention provides a method for quickly identifying the layered information of sediment pollution in shallow lakes by shallow profile. On the basis of shallow profile scanning and columnar sample pollutant measurement, the linear correlation between the water content of each layered sediment and the pollutant content is added. analysis to quickly identify pollution stratification information in lake sediments through shallow profiling.

The invention provides a method for determining the factors affecting the availability of amino acids in rhizosphere soil. By collecting representative zonal soil, sample pretreatment: sieving, grinding, placing in an artificial greenhouse to cultivate plants, adding labeled amino acids, soil Plant samples were processed to measure the physical and chemical properties of the soil, and linear correlation analysis was performed with the amino acid absorption to study the main soil factors affecting the bioavailability of soil amino acids. The method of the invention is simple to operate, and the factors that significantly affect the bioavailability of amino acids can be found among complex factors through correlation detection, which provides a method basis for researching and regulating the bioavailability of amino acids.

The invention provides a method and device for clustering the states of a blast furnace. The method comprises the steps that process parameters of the states of the blast furnace are collected and preprocessed; state feature data of the process parameters is extracted; the state feature data of the process parameters is subjected to linear correlation analysis, and the linear correlation result between the state feature data is determined; when it is determined that the linear correlation result shows linear independence, a K-means algorithm is utilized for clustering the state feature data, and the clustering result of the states of the blast furnace is acquired; and thus, the states of the blast furnace are precisely divided according to the clustering result, and correlation analysis of operation and indexes can be carried out conveniently so that a blast furnace superintendent can grasp the current state of the blast furnace to precisely guide blast furnace operation.

The invention discloses a method and a device for quantitatively distinguishing massive sandstones of the main body of a phase control sand body structure. The method comprises: dividing small layersand taking the comprehensive logging interpretation result as the object of the sand body structure research; performing linear correlation analysis on specified parameters and shale content, selecting top N items with better correlation as the basis of the comprehensive index method, and obtaining the threshold value of the specified parameters according to different log series values and core measured data corresponding to massive sandstones and bedding sandstones; determining a weight coefficient corresponding to each specified parameter; performing data standardization processing on top Nspecified parameters, and calculating the comprehensive index for the parameters after the data standardization processing, in which a weight based on the shale content is used as a coefficient and the directly / inversely proportional relationship of electrical characteristic of the massive sandstones is used as a basis; calculating the smoothness of the comprehensive index through a variance formula and a variation formula. According to the method and the device for quantitatively distinguishing the massive sandstones of the main body of the phase control sand body structure, the massive sandstones can be accurately identified to guide the fine exploration and development of lithological oil-gas reservoir.