11 results about "Grey correlation analysis" patented technology

The present application relates to the technical field of material forming prediction, in particular to a real-time regulation and early warning method in a composite material forming process, comprising acquisition of unstable production composite material morphology images and determination of corresponding processes, analysis of process factors corresponding to each unstable production result by using a grey correlation analysis method, establishment of an early warning model based on a large amount of data, real-time acquisition and analysis of the morphology in actual production, and repeated adjustment of the corresponding process when abnormal production occurs. The model can automatically identify unstable production conditions in the composite material forming process, quickly locate the problem source according to the weight of the process factors, provide early warning information, so as to timely adjust the production parameters and ensure stable production of the composite material.

The application provides a processing parameter optimization method based on a GRA-CRITIC-TOPSIS method and processing surface integrity multi-indexes, and belongs to the technical field of high-efficiency precision processing. In view of the problems that the influence of processing parameters on workpiece surface integrity multi-indexes is complex in difficult-to-machine material milling processing, and the expected results of various indexes are contradictory, the grey correlation analysis method (GRA), the CRITIC objective weighting method and the ideal point method (TOPSIS) are organically combined, and a more scientific and comprehensive multi-objective processing parameter comprehensive optimization model is constructed. The method introduces the CRITIC objective weighting, the weight calculation is more reasonable, the index information overlap can be effectively avoided, and the optimization result has higher engineering reliability.

The present application relates to the technical field of intelligent decision-making, in particular to an engineering change intelligent identification and influence analysis method and system. In the present application, by extracting the physical connection information of the field control equipment and associating the operating parameters and the logical constraint parameters, a field device interconnection topology model with signal flow direction as the connection line is constructed, the discrete engineering equipment is converted into a digital map with logical handshake relationship, the parameter modification behavior is monitored in real time and the change source equipment is locked, the history trend consistency and real-time availability of the redundant signal channel are verified by using the grey correlation analysis algorithm and the effectiveness threshold, when it is confirmed that it has effective backup, it is automatically determined as a blocking node, the input signal change is simulated and calculated according to the PID control algorithm combined with the loop gain coefficient, whether the fluctuation occurs physical dissipation is confirmed by comparing the absolute value of the difference of the theoretical output signal and the signal control dead zone threshold, and the effective fluctuation which can actually drive the downstream equipment action is accurately identified.

The application discloses a target attack priority ranking method and device based on intelligent evaluation, equipment and a medium, and relates to the technical field of data processing.The method comprises the following steps: a target attack priority ranking evaluation index system is constructed; the evaluation index system comprises first-layer indexes and second-layer indexes; the first-layer indexes at least comprise category-type indexes, and the second-layer indexes are used for predicting and evaluating the category-type indexes in the first-layer indexes; a target evaluation decision matrix is constructed based on the first-layer indexes, wherein the values of the category-type indexes in the first-layer indexes are calculated by using an intelligent evaluation method based on a neural network; the weights of the first-layer indexes are calculated by using an entropy weight method; the closeness degrees of the targets are calculated by using a TOPSIS method; the comprehensive correlation degrees of the targets are calculated by using a grey correlation analysis method; the comprehensive evaluation values of the targets are weighted and calculated based on the closeness degrees and the comprehensive correlation degrees; and the targets are ranked according to the comprehensive evaluation values. The application greatly improves the scientificity, objectivity and reliability of the target ranking result.

The application relates to a data trader credit determination method and device, computer equipment, a storage medium and a computer program product. It relates to the technical field of big data. The method comprises the following steps: based on a pre-established evaluation index system, a combination weighting method is used to fuse the subjective weight and the objective weight of each to-be-evaluated index respectively, and the comprehensive weight of each to-be-evaluated index is determined; credit evaluation data of a plurality of data traders is obtained based on a grey correlation analysis method, and an evaluation index data column corresponding to the plurality of credit evaluation data is constructed; based on the to-be-evaluated data corresponding to the data trader, the grey correlation coefficient of the to-be-evaluated data of the trader and a pre-determined reference data column is obtained, and the grey correlation degree of the to-be-evaluated index is obtained in combination with the comprehensive weight; based on the grey correlation degree, the to-be-evaluated index is sorted to determine the credit of the data trader. The method can determine reasonable evaluation indexes in the data transaction process to determine the credit of the data trader.

The application discloses a mixed driving proton exchange membrane fuel cell degradation prediction method, comprising the following steps: obtaining monitoring data; constructing a multivariate grey model; performing grey correlation analysis and selecting data sequences with strong correlation with output voltage sequences; performing Gaussian transformation; obtaining prediction values based on the grey multivariate model; obtaining the predicted output of the proton exchange membrane fuel cell by inversely transforming the prediction values through the Gaussian transformation; constructing a voltage recovery model; modeling the cell stack voltage through polarization parameters, thereby establishing a time-dependent degradation model; introducing a nonlinear recovery framework to utilize the multivariate grey model to establish the voltage recovery model; and obtaining the mixed driving proton exchange membrane fuel cell degradation prediction based on the voltage recovery model and the multivariate grey model. The application focuses on breaking through three major problems, namely, dynamic health index failure, degradation-recovery mechanism coupling and online prediction real-time performance, and realizes high-precision, low-cost and real-time prediction of the degradation trajectory.

A kind of power station auxiliary power state evaluation method based on ANP and grey correlation analysis, comprising: first, build the power station auxiliary power index system covering equipment state, operation and maintenance, safety risk and other multidimensional;Then, using ANP network analytic hierarchy process, combined with expert scoring and nine times scale method to construct judgment matrix, through supermatrix, limit supermatrix calculation reflects the importance of index itself and the objective weight of intergroup feedback;Then collect multi-source data and carry out standardization processing, repair missing abnormal value;Then use grey correlation analysis method to construct correlation coefficient matrix, quantitatively process data grey characteristics;Finally, calculate the weighted correlation degree, and determine the system safety level according to the maximum correlation degree rule.The method significantly improves the objectivity of weight distribution and the data anti-interference ability, and can provide reliable basis for power station auxiliary power system operation and maintenance decision and fault early warning.

The application discloses a precision control method and system of a precision straightening machine based on a fuzzy algorithm, and relates to the technical field of automatic straightening control of shaft parts, rods and strips. When the precision straightening machine is used for multi-point step-by-step straightening of long shaft rods, shaft parts, finished rods or strips, the current straightening point jump amount and parameters such as previous straightening, material hardening, original bending shape, supporting state, downstroke and working load are collected, compensation is generated through feature decomposition, grey correlation analysis and fuzzy reasoning, the downstroke and working load are corrected, and the jump offset data of the non-pressed area and the straightened area are fed back. Through the combination of the coupling correlation distribution sequence and the fuzzy reasoning result, the downstroke and working load are compensated and corrected, the precision straightening machine can take into account the current point and the surrounding area, the repeated adjustment of the straightening point is reduced, and the control fitting degree of the multi-point step-by-step straightening process is improved.

The application discloses a kind of special event's unmanned aerial vehicle transport network local reconstruction method, belong to unmanned aerial vehicle transport network design technical field, including the steps are as follows: by remote sensing, communication equipment, obtain unmanned aerial vehicle transport network information data, the topological structure of transport network is modeled;Through the analysis of two special events of flight restriction, temporary task, the influence range of special event to unmanned aerial vehicle transport network is identified;Unmanned aerial vehicle transport network local reconstruction strategy is formulated, including original network reconstruction strategy and topological structure adjustment strategy;Transport network structure and operation index set are established, and the performance of unmanned aerial vehicle transport network is evaluated by grey correlation analysis method.The application considers dynamic environment influence, by the local reconstruction of unmanned aerial vehicle transport network, the adaptability of transport network is enhanced, and the unmanned aerial vehicle transport network with flexibility and sustainable development is formed.

The application discloses a hydrogen-containing comprehensive energy system optimal operation method considering seasonal scheduling, and specifically comprises the following steps: step 1, a comprehensive energy system integrating a carbon capture power plant and hydrogen energy multi-element utilization is built, and an NBRO-Kmeans++ algorithm is used to select representative typical scenes from annual data; step 2, on the basis of step 1, a scientific carbon emission quota allocation model is established by integrating multiple principles and multiple indexes through a distance between optimal and poor solutions method combined with grey correlation analysis, and carbon emission characteristics in different seasons are analyzed; step 3, on the basis of step 2, when the PIES is subjected to low-carbon economic scheduling, optimal scheduling is carried out with the minimum total operation cost as the target. The application can realize minimization of the operation cost and carbon emission of the comprehensive energy system.