13 results about "Principal component regression" patented technology

The application relates to the technical field of power energy consumption prediction, in particular to a power energy consumption prediction method and system based on multidimensional data and a medium, the method comprising the following steps: collecting power consumption data and various influence data of a power grid in real time; determining redundancy imbalance degrees of the various influence data in a time sequence interval; calculating mutual information dependency degrees of the various influence data in the time sequence interval; identifying each mutation point of the power consumption data in the time sequence interval; determining prediction effective contributions of the various influence data in the time sequence interval; combining the redundancy imbalance degrees and the prediction effective contributions to correct regression coefficients in a prediction process of the power consumption data of the power grid by using a PCR principal component regression algorithm. Therefore, the prediction accuracy of power energy consumption is improved.

A device and method for detecting the gas content of a two-phase flow of oil and gas using a halogen lamp light source are disclosed. The method includes the following steps: Step 1: Turn on the halogen lamp light source and inject the prepared two-phase flow of oil and gas to be tested into the pipeline through a three-way valve according to the pre-calibrated gas content. Step 2: Turn on the peristaltic pump to uniformly fill the pipeline with the two-phase flow of oil and gas to be tested, then turn off the peristaltic pump. A first collimating lens is connected to the halogen lamp light source via a first optical fiber to receive the incident light emitted by the halogen lamp light source. A spectrometer is connected to the second collimating lens via a second optical fiber to receive the outgoing light of the two-phase flow of oil and gas passing through the flow cell. The spectrometer records the spectral information in a computer. Step 3: Repeat steps 1 and 2 until the spectral information of all calibrated gas contents is obtained. Step 4: Establish a prediction model using the spectral information. Specifically, a principal component regression (PCR) model for detecting the gas content of the two-phase flow of oil and gas is established to predict the gas content of the two-phase flow.

The present application relates to a kind of zirconium oxide humidity sensor low pressure compensation method, the method includes the following steps, step 1: first need to the current calibration of sensor, ensure the consistency of sensor, the current of sensor in different states is measured and calibrated to target current;Step 2: measure current pressure or user input current pressure P, standard atmospheric pressure atm is calculated to obtain oxygen vector [K,B];Step 3: the coherence of humidity and oxygen is analyzed using PLSR algorithm, and the principal component regression of PLSR algorithm is obtained to exist interference parameter, and more parameter is used to fit PLS model coefficient [A,B,C,D,E] with pressure;Step 4: oxygen concentration and humidity are calculated by the vector obtained in step 2 and 3;Step 5: in different pressure environment, the same coefficient can also be measured accurately oxygen concentration and humidity;Step 6: when sensor batch changes, only current calibration is needed, and the vector of step 2,3 is also applicable.

The invention relates to the technical field of electric power consumption prediction, in particular to an electric power consumption prediction method and system based on multi-dimensional data and a medium, and the method comprises the steps: collecting power consumption data of a power grid and various influence data in real time; determining the redundancy unbalance degree of each influence data in the time sequence interval; calculating mutual information dependency degrees of various influence data in the time sequence interval; identifying each abrupt change point of the electricity consumption data in the time sequence interval, and determining the prediction effective contribution of each influence data in the time sequence interval; and in combination with the redundancy unbalance degree and the prediction effective contribution, correcting a regression coefficient in a process of predicting the power consumption data of the power grid by using a PCR principal component regression algorithm. Therefore, the prediction precision of the power consumption is improved.

The invention discloses a chemical reaction organic matter real-time detection system based on a near-infrared instrument, which comprises the following steps: collecting a spectral signal of a reaction sample under irradiation of near-infrared light, carrying out standardization treatment on original data, extracting nonlinear characteristics by adopting an improved kernel principal component regression method, and calculating a nonlinear characteristic of the reaction sample; a Dirichlet process hybrid model is combined to realize adaptive clustering, then a posterior probability weight strategy is constructed to realize dynamic prediction of the concentration of the organic matter, finally detection output data with time stamps are generated, and real-time monitoring and closed-loop modeling control of organic matter components in chemical reaction are realized. The method is used for supporting component monitoring and control in the chemical process.

The invention relates to a cable degradation degree detection method and device based on principal component regression. The method comprises the following steps: acquiring a plurality of harmonic data in currents of a plurality of circuits, and performing data conversion on the plurality of harmonic data to obtain a plurality of harmonic data in a target format; performing data preprocessing on the harmonic data in the target format to obtain preprocessed target harmonic data; performing principal component analysis on the target harmonic data to obtain a data matrix corresponding to the target harmonic data; inputting the data matrix corresponding to the target harmonic data into a pre-trained target principal component regression model to obtain a cable degradation degree prediction value of the circuit corresponding to the target harmonic data; determining the degradation state of the cable of the circuit according to the cable degradation degree prediction value of the circuit; and determining a circuit cable maintenance suggestion according to the degradation state of the circuit cable. Stable and accurate degradation degree detection can be performed on the cable at least through a principal component regression model and a principal component analysis method.

The invention relates to the technical field of precision control of numerical control machine tools, and discloses an improved principal component regression machine tool thermal error modeling method based on density clustering, which comprises the following steps of: performing dimension reduction processing on original temperature data through principal component analysis to obtain a plurality of principal components; establishing a plurality of principal component regression models based on the principal components, wherein each model corresponds to a different principal component number; performing clustering analysis on prediction results of the plurality of principal component regression models by using a density clustering algorithm, and identifying and eliminating abnormal prediction values; and performing comprehensive processing on the prediction result after the abnormal value is removed to obtain a final thermal error prediction model. The principal component regression model prediction results corresponding to different principal component numbers are screened and optimized by combining a density clustering algorithm, abnormal prediction values are effectively eliminated, and reliable prediction results are integrated, so that a thermal error prediction model with higher precision is constructed.

The invention provides a method for constructing a prediction model for early warning of acquired myasthenia related to sepsis based on an electrophysiological index, a prediction method and application, and the construction method comprises the steps: determining the electrophysiological index most related to an APACHE II score and an SOFA score based on factor analysis and principal component regression analysis; the method comprises the following steps: acquiring personal information, electrophysiological indexes, inflammatory response and score, MRC score and EMG, RNS and DMS measurement data of a sample SIRS patient; constructing different attribute sets according to data categories, and constructing a similarity fusion network model based on the attribute sets; and based on a similarity measurement matrix output by the similarity fusion network model, carrying out clustering analysis according to a spectral clustering algorithm, outputting a first classification category and a confusion matrix, and based on the confusion matrix, taking the similarity fusion network model meeting accuracy and precision requirements as an early prediction model of acquired myasthenia related to sepsis. The prediction model can predict the type of the patient, and is beneficial to early intervention and patient rehabilitation.

The invention provides a short-term power load interval prediction method based on kernel principal component regression analysis. The method comprises the following steps: acquiring historical power load data and corresponding historical power load influence data; clustering the historical power load influence data, and determining the historical power load influence data corresponding to each power consumption scene according to a clustering result; for the historical power load influence data corresponding to each power consumption scene, performing dimension reduction based on kernel principal component analysis, and extracting main power load influence characteristics corresponding to each power consumption scene; constructing a load interval prediction model based on the main power load influence characteristics corresponding to each power consumption scene and the corresponding historical power load data; and performing interval prediction on the short-term power load of the power system according to the load interval prediction model corresponding to each power consumption scene. The method can provide an interval prediction result for short-term power load prediction of the power system, overcomes the uncertainty of the prediction result, and gives consideration to the prediction efficiency.

The invention relates to a quality-related fault detection method based on a deep decomposition echo state network, which comprises the following steps of: dividing into process variable data and quality variable data, and standardizing by using a mean value and a variance of the process variable data and the quality variable data; the method comprises the following steps: constructing a first decomposition network to extract residual information, decomposing a process variable matrix U by using a principal component regression method, sending the decomposed process variable matrix U into a deep echo state network to extract potential residual information, calculating residual vectors # imgabs0 # and # imgabs1 # to construct a second decomposition network to extract dynamic information, and extracting dynamic characteristics X by using an echo state network method. Decomposing the X by using a principal component regression method, further sending the X into a deep echo state network to extract dynamic information in the X, and respectively calculating score matrixes Tre and Tun of the X; fusing and constructing a quality-related statistic # imgabs2 # and a quality-independent statistic # imgabs3 #, and calculating the control limits of the quality-related statistic # imgabs2 # and the quality-independent statistic # imgabs3 And collecting real-time process data, performing standardization processing, calculating a residual vector and a score vector of online data, and judging whether a fault affects a quality variable or not according to a statistical magnitude detection result. According to the fault detection method, the echo state network and the principal component regression method are combined, and the detection performance of industrial process data quality related faults is improved.

The invention belongs to the technical field of soil heavy metal spectrum determination, and particularly relates to a method for rapidly determining the content of low-concentration heavy metal in grassland soil around a mining area based on visible light-near infrared spectroscopy. In order to realize rapid determination of the heavy metal content based on hyperspectral data, three statistical methods of partial least squares regression (PLSR), principal component regression (PCR) and support vector machine regression (SVMR) and 16 preprocessing combinations are researched, developed and explored, the optimal combination is screened out, the determination mechanism of the heavy metal content is analyzed by the methods, and the determination result of the heavy metal content is obtained. The correlation between the heavy metal content in the soil around the mining area and spectral analysis is defined, and meanwhile, different pretreatment methods and statistical methods are compared, so that an important scientific support is provided for heavy metal pollution research. According to the method, variable information can be enhanced, model errors are reduced, and the accuracy and stability of the model are improved.

The application discloses a large white pig genome prediction method based on improved genome feature principal component regression, which is defined as GF_PCR. The method divides the traditional PCR method into two parts, performs simple regression on the genome feature part, and performs PCR analysis on the remaining SNPs. When the genome feature contains more causal mutations, the model can obtain higher genome prediction accuracy. At the same time, the genome prediction accuracy of the GF_PCR model for the 100Kg body weight day age trait is higher than that of the PCR model, and the reason is that the proportion of the causal mutation in the preselected genome feature is larger, and the GF_PCR further improves the genome prediction accuracy.

The application discloses a kind of urban space net red phenomenon determination method, it is related to urban space technical field, including the following steps: collection short video check-in data, obtain the online traffic and offline traffic of multiple net red check-in points;Based on the ratio of offline traffic and online traffic, obtain the traffic conversion rate of multiple net red check-in points in the preset research range;The surrounding area of net red check-in point is researched, and the influence factor of urban space net red phenomenon is determined and corresponding data is obtained;Based on the traffic conversion rate of net red check-in point and each influence factor, linear regression model is established respectively, and multiple regression analysis equation is obtained;The weight of each influence factor is obtained by using principal component regression analysis method, and weight analysis result is obtained.The application can analyze the correlation between urban net red phenomenon and influence factor, help to develop urban space potential, and help to promote the development of urban space net red phenomenon by optimizing space quality targetedly.