9results about How to "Good prediction" patented technology

The invention discloses a multi-axis notched part fatigue life estimation method based on an energy criterion, and the method comprises the steps: firstly, giving a calculation method of equivalent strain energy density, and representing an energy distribution condition on a critical surface through a hexa-degree polynomial; determining a critical distance by using the relative energy gradient, giving a weight function based on the energy gradient, and calculating strain energy density on a critical surface; secondly, giving a fatigue life prediction model of the multi-axis notched part by considering the influence of energy gradient, size effect and non-proportional additional strengthening effect; the prediction result of the model provided by the invention is within 2 times of the dispersion zone, and is superior to the prediction result of other models.

The invention discloses a three-dimensional wind turbine wake flow calculation method considering a yaw effect, which comprises the following steps of: calculating a flow direction speed distribution profile u0 (z) and a turbulence intensity distribution profile I0 (z) of inlet wind according to a basic rule of turbulence motion in a boundary layer; calculating an axial flow induction factor a of the wind turbine in a yaw state, and then determining a wake flow expansion radius rx at the downstream x position of the wind turbine; calculating the additional turbulence intensity Iadd, x at the position x of the wake flow area; the wake flow center offset yd (x) is calculated, and then the radiation radius r with the wake flow center as the circle center is determined; judging the wake flow area of the wind turbine, and calculating the average wake flow speed u * (x, z) of the wake flow area at the position where the vertical direction height is z at the downstream x distance of the wind turbine; and calculating to obtain the wind turbine airflow area speed distribution. According to the method, the wake flow center offset and wake flow speed distribution characteristics of the yaw wind turbine can be accurately predicted.

The invention discloses a blue-green algae growth prediction method and an intelligent regulation and control device, and relates to the field of blue-green algae growth prediction and regulation and control, and the method comprises the steps: carrying out the collection of blue-green algae growth related data, data preprocessing, random forest model construction, SVM model construction, and model weight fusion of a final random forest prediction model and a final SVM prediction model to obtain a fusion model. And dynamically optimizing the fusion model, finally obtaining the dynamically optimized fusion model, obtaining a real-time prediction result according to the dynamically optimized fusion model by using the blue-green algae growth related data collected by the sensor module in real time, and adjusting the blue-green algae growth culture conditions in real time. According to the method, the advantages of the random forest model and the SVM model are combined, a fusion model with dynamic optimization and real-time adjustment functions is developed, and the accuracy and adaptability of blue-green algae growth prediction are remarkably improved. And meanwhile, data acquisition, processing and prediction modules are integrated, so that the accuracy of blue-green algae growth prediction and the automation degree of the system are remarkably improved.

The invention discloses a method for solving the formula proportion of each component oil in a blended oil product based on near infrared spectroscopy, which comprises the following steps: collecting the component oil participating in blending, firstly carrying out spectrum scanning on the pure component oil through a near infrared spectrum analyzer, and respectively obtaining near infrared spectrum data X1, X2... Xn of each component oil; the method comprises the following steps: performing spectrum scanning on blend oil with different formula proportions through a near infrared spectrum analyzer, collecting corresponding near infrared spectrum data Y1, Y2... Yp, pre-processing the near infrared spectrum of the mixed oil product by setting a solving target and constraint conditions, and quickly solving the formula proportion of the mixed oil product by utilizing an lsplin optimization algorithm. The method can be used for the online monitoring process of oil products, is good in robustness, and can still accurately predict the formula ratio of each oil product even in the mixing process of various oil products.

The present invention provides a shear wave prediction method for organic-rich shale based on the Xu-white model with a variable pore aspect ratio, the method comprising the following steps: according to the characteristics of kerogen, it is simultaneously equivalent to matrix minerals and pore fluids; based on the Xu-white model, the Voigt-Reuss-Hill average theory is introduced to calculate the equivalent elastic modulus of the organic-rich mineral matrix, the Wood formula is used to calculate the equivalent bulk modulus of the mixed fluid, the Gassmann fluid substitution equation is used to calculate the bulk modulus and shear modulus of the fluid-saturated organic-rich shale, and the density, porosity, mud content and longitudinal wave velocity of the fluid-saturated shale rock are used as the variable pore aspect ratio to construct the Xu-White model with a variable pore aspect ratio, and then predict the shear wave velocity. The Xu-White model constructed by the present invention can be effectively applied to the prediction of shear wave velocity of organic-rich shale reservoirs, takes into account the special elastic properties of kerogen, enhances the applicability of the model, avoids the error caused by the rock physical parameters being constants, and improves the accuracy of the prediction results.

A method for predicting compound hERG cardiotoxicity based on multidimensional molecular fingerprints, the method comprising: the first step, performing molecular dynamics simulation on compounds in a cardiotoxicity data set; the second step, analyzing the molecular dynamics simulation results and calculating molecular fingerprints based on molecular dynamics simulation; the third step, calculating the molecular fingerprint of the compound by molecular fingerprint software, and forming a multidimensional molecular fingerprint with the molecular dynamics simulation fingerprint; the fourth step, using feature selection methods to find the optimal feature subset from the multidimensional molecular fingerprint; the fifth step, using multiple machine learning methods to construct a compound hERG cardiotoxicity prediction model. Experimental results show that the model established using multidimensional molecular fingerprints has higher prediction performance. This method can achieve higher compound hERG cardiotoxicity prediction accuracy and can be used for early toxicity screening in drug design.

The invention belongs to the technical field of marine science and data processing, and provides a base model construction method and system for marine observation data, and the method comprises the steps: constructing a marine base model; inputting the acquired ocean time series data into an ocean foundation model, constructing a Token sequence of a timestamp granularity level of the ocean time series data, and connecting an offset delay feature vector and a covariable vector before a timestamp into a timestamp Token vector; in combination with a causal self-attention mechanism and rotation position embedding, the features of the Token sequence are mapped to a decoder of a Transform, and a next timestamp sequence is generated; and defining a loss function, and optimizing model parameters to obtain a trained ocean foundation model. According to the method, universal feature representation and time sequence modes are learned on various ocean data sets, so that an ocean universal model with high generalization ability is constructed, large-scale and multi-source ocean data can be processed, time sequence characteristics of the data can be captured, and real-time data processing and prediction ability can be provided.

The invention discloses a notched part fatigue life prediction method based on energy gradient. The notched part fatigue life prediction method comprises the steps of determining a critical plane based on an energy method; determining a size effect influence factor based on the energy gradient and the notch geometric parameters; and determining damage parameters. According to the method, the concept of strain energy gradient factors is proposed to describe complex non-uniform energy distribution around the notch, and the influence of stress and strain changes on the fatigue life is well reflected; quantifying different contributions of shear and normal strain energy to damage by using an energy ratio on a critical surface; and introducing a size effect factor to quantify the influence of the size effect on the fatigue life of the notched test piece by combining an energy gradient factor and a notched size parameter.

The present application relates to the technical field of lithium ion battery, and provides a GRU-EC-GAPSO lithium ion battery SOH prediction method, which comprises the following steps: extracting a health factor from voltage and time data in a local SOC interval during charging and discharging of a lithium ion battery, inputting the GRU model to perform SOH prediction, calculating an error sequence of a predicted value and an actual value, training an EC model by using the error sequence, and obtaining a GRU-EC model; using a GAPSO algorithm to optimize parameters of the GRU-EC model, and obtaining a GRU-EC-GAPSO model; and applying the trained GRU-EC-GAPSO model to different types of lithium ion battery data to perform SOH prediction. The present application has high prediction accuracy and good robustness in lithium ion battery SOH prediction.