272results about How to "Improve predictive ability" patented technology

Genes and gene expression profiles useful for predicting outcome, risk classification, cytogenetics and/or etiology in pediatric acute lymphoblastic leukemia (ALL). OPAL1 is a novel gene associated with outcome and, along with other newly identified genes, represent a novel therapeutic targets.

The invention relates to a method for predicting a low-porosity and low-permeability clasolite effective reservoir at a high-diagenesis stage and belongs to the technical field of oil and gas exploration and development. The method is characterized in that a geological process response relationship is established through carrying out analysis, simulation and quantitative evaluation on a geological process affecting porosity, the original pore preservation and induced pore increase/decrease conditions of a reservoir are determined, thus, the current porosity of a low-porosity and low-permeability reservoir at the high-diagenesis stage is quantitatively evaluated, the development of the effective reservoir is predicted, and a basis is provided for reservoir evaluation. The method for predicting the low-porosity and low-permeability clasolite effective reservoir at the high-diagenesis stage has the advantages that the predictability is strong, the distribution of various clasolite effective reservoirs at different stages of high-diagenesis can be accurately predicted, the fine prediction on the distribution of effective reservoirs of a research target section can be achieved, and the predicted unit plane can reach a hectometer level; the theoretical basis is solid, and the condition of over reliance on the data of sampled points in the traditional method is avoided; the quantitative performance is good, the applicability is wide, and the information requirements are relatively low; and the evaluation, with different information foundations, on different exploration and development stages is facilitated.

The invention discloses an intelligent maintenance system and method based on a wearable augmented reality display terminal and a cloud platform. According to the intelligent maintenance system and method, an augmented reality technology-based infrared image intelligent identification, video collection and uploading of personnel operation scenarios and real-time conversation with a monitoring center on the scene are achieved, thereby guiding maintenance personnel to implement standard and normalized work process; an intelligent maintenance platform based on big data and artificial intelligenceis built to carry out inspection optimization, defect identification and fault diagnosis; and auxiliary information is transmitted to an intelligent terminal of site operation personnel to display inreal time through an augmented reality technology, so that the operation efficiency is improved, misoperation is avoided and the intelligent maintenance system and method can be applied to multiple power grid operation and maintenance scenarios.

Systems and methods are provided for handling time-stamped data. The one or more GUIs may be used to define properties of an event and to associate the event with time-stamped data. The properties of the event defined using the one or more GUIs may include one or more statistical properties that indicate how the event statistically affects the time-stamped data. A time series analysis program may use the time-stamped data and the associated event to generate a mathematical model.

The invention discloses a 3D visualized smart water affairs operation system. The system comprises a sewage plant automatic control module, a sewage plant optimization module, a pumping station automatic control module, a water quality monitoring module, a plant area display module, a video monitoring module, an equipment management module, an energy consumption management module, an intelligent tour inspection module, a pipe network management module, a data analysis module and a mobile end module connected with a central database via a central scheduling module; the video monitoring module is connected with the data analysis module; and the data analysis module is connected with the central database and the central scheduling module. According to the system, a 3D virtual reality technology is used, a complete remote monitoring, early warning, scheduling, disposal and control system is established, an intelligent integrated command platform and mobile tour inspection equipment are constructed in an integrated manner, the capability of coping with change is improved, operation management is made digital, informational, modern, automatic and intelligent, remote data monitoring and control are realized, and the management efficiency and precision are improved greatly.

The invention belongs to the field of photovoltaic power prediction, and discloses a CNN-based LSTM photovoltaic power generation power prediction model and a construction method. The model onsists ofa two-layer CNN convolutional layer, an LSTM, an attention mechanism and a full connection layer. When only power data exists, feature extraction is carried out on an input one-dimensional power generation power time sequence through a double-layer convolution layer, a plurality of data matrixes are obtained through 3 * 1 convolution kernel operation. Information abstraction from the time sequence to high-level features is achieved. The attention mechanism automatically pays attention to an LSTM hidden layer output vector obtained by each matrix through an LSTM algorithm, and a greater weightis given to a characteristic quantity remarkably related to the current output quantity. The output vector of the attention mechanism is processed into a one-dimensional vector through an unfolding operation and inputted the one-dimensional vector into the full connection layer, wherein a full connection layer directly outputs the predicted value of the photovoltaic power generation power at thenext moment. The method is small in data size, high in transportability and high in prediction precision.

The invention discloses a target image identification transmission line state monitoring system based on a visual attention mechanism. The target image identification transmission line state monitoring system comprises a monitoring device on a tower pole, another monitoring device on a line and a monitoring center, wherein the monitoring device on the tower pole comprises a data image acquisition terminal, a master control unit and a wind-solar complementary system; the monitoring device on the line comprises an acquisition terminal on the line, a master control unit on the line and a current sensing electricity taking device; and both the monitoring device on the tower pole and the monitoring device on the line send the data to the monitoring center by a wireless video and data communication unit via a GPRS (General Packet Radio Service)/EDGE (Electronic Data Gathering Equipment)/3G network. The whole system is a comprehensive application system which integrates software and hardware, as well as integrates automatic control, computer data processing and communication, the purpose of all-weather on-line monitoring under any complex weather condition is realized, and the image identification efficiency and the transmission line state detection accuracy are substantially improved.

Disclosed is a protein-ligand affinity predicting method based on molecule descriptors. The protein-ligand affinity is reflected through construction of perfect and systematic molecule descriptors, and the relation between the descriptors and the affinity is constructed through a supporting vector regression (SVR) mode. The method includes the steps of training set preparation: preparing a large amount of data containing the crystal structure and the affinity of a protein-ligand complex; construction and calculation of the molecule descriptors: constructing 50 kinds of molecule descriptors which belong to nine categories, and calculating concrete values of all the complex descriptors in the training set; regression model construction: fitting the relation between the descriptors and the affinity through the SVR mode, and introducing a conjugate gradient method to optimize a penalty factor C and a kernel function parameter; and novel scoring function building which is used for predicting the affinity of the complex. The method has the advantages of being high in prediction capacity, small in target dependence, high in homolog sensitivity and the like.

The invention belongs to the field of medicine preparations, which in particular relates to a method for distinguishing the coating quality. More concretely, the invention relates to a method for distinguishing the coating quality of a medicine preparation. In the method provided by the invention, a qualitative distinguishing model of the coating quality is quickly constructed by combining the methods of a near-infrared spectrum method, an interval principal component analysis method, a support vector machine method and the like, the comprehensive evaluation of the coating quality of the medicine preparation is realized, and the distinguishing accuracy basically reaches the requirement of industrial production. Relative to a traditional method or experiential judgment, the distinguishing method provided by the invention can quickly, nondestructively, reliably and accurately distinguish the coating quality of the medicine preparation in a coating process in an on-line way.

The invention relates to an infrared spectrum recognition method for adulterated milk. The method comprises the following steps that pure milk, urea-doped milk and glucose-doped milk are taken as objects of study respectively, an infrared spectrogram of a sample is collected by utilizing an ATR-FTIR (attenuated total reflection-fourier transform infrared spectroscopy) technique, and adulterants are recognized qualitatively and analyzed quantitatively through establishing SIMCA (Soft Independent Modeling Of Class Analogy) and PLS (Partial Least Squares) models. According to the invention, when the concentration of the adulterants of urea and glucose is 0.10-1.00%, the recognition rate of the adulterants of urea and glucose by the method is 100%. When the concentration of the adulterants is 0.01-0.10%, the recognition rate and the rejection rate of the pure milk and the milk doped with urea and glucose by a best model are both greater than 82%, and the effect is favorable. The relevance of a prediction value and an actual measured value of a model is favorable for the adulterants of urea and glucose; the established models have favorable prediction capability, and can be used for the quantitative analysis of multiple adulterants in milk.

The invention discloses a quick edible oil variety identification method based on multisource spectral feature fusion. The method comprises the steps of selecting an unknown variety of edible oil sample to be identified, acquiring a Raman spectrum chart and a near infrared spectrum chart, preprocessing the Raman spectrum chart and the near infrared spectrum chart to obtain a preprocessed Raman spectrum chart and a preprocessed near infrared spectrum chart, conducting feature extraction on the preprocessed Raman spectrum chart and the preprocessed near infrared spectrum chart to obtain a Raman characteristic variable and a near infrared characteristic variable, conducting spectral characteristic fusion on the Raman characteristic variable and the near infrared characteristic variable to obtain a characteristic fusion spectrum chart, and conducting variety identification on the unknown variety of edible oil sample by means of an optimized qualitative model. The method is safe and quick, detection is convenient, identification accuracy is high, and practical value and popularization value are high.

The invention provides a screening method of all inherited metabolic disorder genes for genetic diagnosis within an exon area, which is fast and accurate and can cover the newest inherited metabolic disorder genes. The method provided by the invention comprises the following steps of: drawing 3-5ml of blood from an individual, extracting 3-5 microgrammes of DNA (Deoxyribonucleic Acid) from the blood, interrupting and amplifying the DNA to construct a whole genome library for the patient, capturing the virulence genes by using an inherited metabolic disorder gene scanning kit provided by the invention, carrying out high-throughput sequencing by using a sequencing machine, and analyzing and finding mutation information relevant to the genes so as to obtain the mutation conditions of the inherited metabolic disorder genes of the individual to reach the purpose of accurate genetic diagnosis. Dozens of to thousands of genes and millions of loci can be captured and detected once by taking advantage of the high-throughput sequencing, and the screening method covers known 700 inherited metabolic disorders.

The invention provides a determination method of a carthamus tinctorius extract. A least squares support vector machine theory and a near infrared spectrum technology are combined, and a particle swarm optimization is used for parameter optimization on LS-SVM, so as to establish a quantitative analytical mode of amount of solid and hydroxyl safflower yellower A during a carthamus tinctorius extraction and realize rapid determination of key indexes of the amount of solid and a hydroxyl safflower yellower A concentration during the safflower extraction. The mode established on least squares support vector machine of particle swarm optimization of the invention has advantages of good stability, high degree of forecast accuracy and strong extrapolation and generalization capability. Compared with a traditional analysis method, the established analysis method is rapid and effective, and can be used for rapid determination of solid amount and hydroxyl safflower yellower A concentration during a carthamus tinctorius extraction. The invention is conducive to increase of quality control level during safflower extraction, so as to ensure stable and reliable product quality. The method of the invention has advantages of simple operation, strong model prediction capability, high stability, and strong extrapolation and generalization capability.

The invention provides an anti-scratch device and method used during vehicle turning. The anti-scratch device comprises two laser scanning radars installed on the left side and the right side of a front bumper of a vehicle correspondingly, a steering wheel angle sensor installed on a steering shaft of a steering wheel of the vehicle, an intelligent CAN converter, an alarm prompt unit and a processing unit. The anti-scratch device and method used during vehicle turning have the beneficial effects that the laser scanning radars are used for obtaining the road environment information around the vehicle, laser radar detection is not disturbed by ambient light, and the recognition precision is high. Meanwhile, a vehicle track is determined according to the angle of the steering wheel, the next position of the vehicle is further determined, and a system has high predictability; and the method is simple, and the advanced active safety system is obtained.

The invention discloses an enterprise relocation risk assessment method based on a logical regression model, which comprises the following steps of S1 collecting the information field of the enterprise relocation instance and the desensitized telecommunication data provided by an operator, and selecting relevant variables of the enterprise relocation; S2 dividing the related variables into data boxes; 3 screening out low prediction ability indexes; S4 eliminating the high correlation index; S5 constructing the enterprise emigration prediction model by logical regression; 6 testing that prediction model; S7 evaluating the risk of the enterprise relocation through the enterprise relocation prediction model outputted from the step S6. The enterprise migration prediction model based on the logical regression fills in the blank of the enterprise migration quantitative analysis model, and has strong prediction ability.

The invention discloses a stress traffic scene driving training system based on a virtual reality technology. The stress traffic scene driving training system generates an emergency traffic scene which can cause stress response of a driver during the vehicle driving process by adopting the virtual reality technology, and truly reproduces the stress traffic scene by means of a helmet-mounted display, a loudspeaker and a force feedback device. The stress traffic scene driving training system obtains operation data during the process of interacting with vision, auditory sense and tactile sense ofthe driver, performs driving skill evaluation on the driver according to the operation data, and establishes dreadful traffic accident scenes, sound effect and strong tactile impact based on error operations of the driver to warn the driver, so as to achieve the purpose of improving the driving level. The traffic accident disclosed by the invention is conductive to improving the stress traffic scene driving skill which is difficult for the driver to obtain in the actual driving process, enhances the prejudging capability of the driver for the stress traffic scene, establishes good stress traffic scene driving habits, improves the road traffic safety level, and has broad application prospects.

The present invention relates to the technical field of water level prediction of ship nuclear power voltage regulators, and particularly relates to a voltage regulator water level prediction method based on a cost-sensitive LSTM cyclic neural network. The voltage regulator water level prediction method based on the cost-sensitive LSTM cyclic neural network comprises the following steps: S1, selecting p parameters (as shown in the specification) with a relatively high degree of coupling with a voltage regulator water level as input parameters; S2, constructing a LSTM-based voltage regulator water level prediction model and a framework thereof; and S3, using a BPTT algorithm to train and optimize the water level prediction model constructed in step S2. According to the method provided by the present invention, the LSTM model can better approximate the true value of the water level than the SVR model and the BP neural network model, and has stronger learning ability and prediction ability, and the cost-sensitive LSTM model has better precision and faster convergence.

The invention relates to a fluorescent PCR detection method for a human cytomegalovirus (HCMV). In the invention, two groups of detecting primers and probes are designed for UL122 and UL134 genes of a human cytomegalovirus (HCMV) genome. By detecting the two genes simultaneously, a false negative rate caused by the mutation of the HCMV genome is reduced. The technology of the invention can perform qualitative and quantitative detection on HCMV DNA quickly and conveniently and has relatively high specificity and sensitivity.

The invention discloses a high-flux detection method of a circulating DNA liver cancer driver gene. According to the method, a mutant gene which is the most common of the existing liver cancer is selected and combined with a second-generation high-flux sequencing technology to more comprehensively detect a gene mutant condition of the liver cancer. Since a blood plasma sample of a patient is adopted, the method can greatly reduce the detection injury of a patient; secondly, the whole-exome detection is performed on 75 liver cancer genes, the method is more comprehensive compared to the detection method based on real-time PCR and first-generation Sanger sequencing technology; and since only 75 genes are detected, the sample can be detected in a greater sequencing depth, so that the detection precision can reach 0.1 percent. Therefore, a reference can be provided for more and better individual precision therapeutic schemes of the patient.

The invention provides a radar echo extrapolation prediction method and system based on a velocity field sensing network. The prediction method includes: inputting radar echo data and velocity field data into a trained velocity field sensing radar extrapolation network model to obtain a prediction radar echo sequence, wherein the trained velocity field sensing radar extrapolation network model isobtained by the following steps of building a memory at any time, building an output gate at any time, obtaining a hidden state at any time, constructing a long short-term memory network structure atany time, building and integrating a radar extrapolation network model with a velocity field to obtain the velocity field sensing radar extrapolation network model, and inputting tensor sequence datainto the velocity field sensing radar extrapolation network model for training to obtain a trained velocity field sensing radar extrapolation network model. According to the invention, information hidden in historical data is mined. The velocity field is combined in the velocity field sensing radar extrapolation network model, which makes the model have more accurate prediction capability.

The invention belongs to the technical field of tire dynamics and discloses a tire side-tipping side-inclining steady-state aligning torque characteristic radius semi-empirical modeling method. The method comprises decomposing aligning torque in the condition of non-linear stacking of the tire side-tipping side-inclining into side-inclining side force aligning torque, side-tipping side force aligning torque and side-tipping longitudinal force aligning torque, and tire side-tipping side-inclining steady-state aligning torque test data are obtained by testing through a tire mechanical characteristic test table, and parameters of the built tire model are obtained by identification through a curve fitting technology. The model has the advantages of being high in accuracy, clear in parameter physical meaning, capable of meeting theoretical boundary conditions and high in predictive capacity and can be used for high-accuracy car dynamic simulation.