Using deep neural networks coupled with principal component analysis for ore production forecasting at open-pit mines

Ore production is usually affected by multiple influencing inputs at open-pit mines.Nevertheless,the complex nonlinear relationships between these inputs and ore production remain unclear.This becomes even more challenging when training data(e.g.truck haulage information and weather conditions)are massive.In machine learning(ML)algorithms,deep neural network(DNN)is a superior method for processing nonlinear and massive data by adjusting the amount of neurons and hidden layers.This study adopted DNN to forecast ore production using truck haulage information and weather conditions at open-pit mines as training data.Before the prediction models were built,principal component analysis(PCA)was employed to reduce the data dimensionality and eliminate the multicollinearity among highly correlated input variables.To verify the superiority of DNN,three ANNs containing only one hidden layer and six traditional ML models were established as benchmark models.The DNN model with multiple hidden layers performed better than the ANN models with a single hidden layer.The DNN model outperformed the extensively applied benchmark models in predicting ore production.This can provide engineers and researchers with an accurate method to forecast ore production,which helps make sound budgetary decisions and mine planning at open-pit mines.

Tool Health Condition Recognition Method for High Speed Milling of Titanium Alloy Based on Principal Component Analysis (PCA) and Long Short Term Memory (LSTM)

The healthy condition of the milling tool has a very high impact on the machining quality of the titanium components.Therefore,it is important to recognize the healthy condition of the tool and replace the damaged cutter at the right time.In order to recognize the health condition of the milling cutter,a method based on the long short term memory(LSTM)was proposed to recognize tool health state in this paper.The various signals collected in the tool wear experiments were analyzed by time-domain statistics,and then the extracted data were generated by principal component analysis(PCA)method.The preprocessed data extracted by PCA is transmitted to the LSTM model for recognition.Compared with back propagation neural network(BPNN)and support vector machine(SVM),the proposed method can effectively utilize the time-domain regulation in the data to achieve higher recognition speed and accuracy.

Primary component analysis method and reduction of seismicity parameters

In the paper, the primary component analysis is made using 8 seismicity parameters of earthquake frequency N (ML≥3.0), b-value, η-value, A(b)-value, Mf-value, Ac-value, C-value and D-value that reflect the characteristics of magnitude, time and space distribution of seismicity from different respects. By using the primary component analysis method, the synthesis parameter W reflecting the anomalous features of earthquake magnitude, time and space distribution can be gained. Generally, there is some relativity among the 8 parameters, but their variations are different in different periods. The earthquake prediction based on these parameters is not very well. However, the synthesis parameter W showed obvious anomalies before 13 earthquakes (MS≥5.8) occurred in North China, which indicates that the synthesis parameter W can reflect the anomalous characteristics of magnitude, time and space distribution of seismicity better. Other problems related to the conclusions drawn by the primary component analysis method are also discussed.

Comprehensive multivariate grey incidence degree based on principal component analysis

To overcome the too fine-grained granularity problem of multivariate grey incidence analysis and to explore the comprehensive incidence analysis model, three multivariate grey incidences degree models based on principal component analysis （PCA） are proposed. Firstly, the PCA method is introduced to extract the feature sequences of a behavioral matrix. Then, the grey incidence analysis between two behavioral matrices is transformed into the similarity and nearness measure between their feature sequences. Based on the classic grey incidence analysis theory, absolute and relative incidence degree models for feature sequences are constructed, and a comprehensive grey incidence model is proposed. Furthermore, the properties of models are researched. It proves that the proposed models satisfy the properties of translation invariance, multiple transformation invariance, and axioms of the grey incidence analysis, respectively. Finally, a case is studied. The results illustrate that the model is effective than other multivariate grey incidence analysis models.

Laser-induced breakdown spectroscopy applied to the characterization of rock by support vector machine combined with principal component analysis

Laser-induced breakdown spectroscopy（LIBS） is a versatile tool for both qualitative and quantitative analysis.In this paper,LIBS combined with principal component analysis（PCA） and support vector machine（SVM） is applied to rock analysis.Fourteen emission lines including Fe,Mg,Ca,Al,Si,and Ti are selected as analysis lines.A good accuracy（91.38% for the real rock） is achieved by using SVM to analyze the spectroscopic peak area data which are processed by PCA.It can not only reduce the noise and dimensionality which contributes to improving the efficiency of the program,but also solve the problem of linear inseparability by combining PCA and SVM.By this method,the ability of LIBS to classify rock is validated.

A novel method for chemistry tabulation of strained premixed/stratified flames based on principal component analysis

The principal component analysis （PCA） is used to analyze the high dimen- sional chemistry data of laminar premixed/stratified flames under strain effects. The first few principal components （PCs） with larger contribution ratios axe chosen as the tabu- lated scalars to build the look-up chemistry table. Prior tests show that strained premixed flame structure can be well reconstructed. To highlight the physical meanings of the tabu- lated scalars in stratified flames, a modified PCA method is developed, where the mixture fraction is used to replace one of the PCs with the highest correlation coefficient. The other two tabulated scalars are then modified with the Schmidt orthogonalization. The modified tabulated scalars not only have clear physical meanings, but also contain passive scalars. The PCA method has good commonality, and can be extended for building the thermo-chemistry table including strain rate effects when different fuels are used.

Watermarking Based on Principal Component Analysis

A new watermarking scheme using principal component analysis (PCA) is described.The proposed method inserts highly robust watermarks into still images without degrading their visual quality. Experimental results are presented, showing that the PCA based watermarks can resist malicious attacks including lowpass filtering, re scaling, and compression coding.

Characterization of three-dimensional channel reservoirs using ensemble Kalman filter assisted by principal component analysis

Ensemble-based analyses are useful to compare equiprobable scenarios of the reservoir models.However,they require a large suite of reservoir models to cover high uncertainty in heterogeneous and complex reservoir models.For stable convergence in ensemble Kalman filter(EnKF),increasing ensemble size can be one of the solutions,but it causes high computational cost in large-scale reservoir systems.In this paper,we propose a preprocessing of good initial model selection to reduce the ensemble size,and then,EnKF is utilized to predict production performances stochastically.In the model selection scheme,representative models are chosen by using principal component analysis(PCA)and clustering analysis.The dimension of initial models is reduced using PCA,and the reduced models are grouped by clustering.Then,we choose and simulate representative models from the cluster groups to compare errors of production predictions with historical observation data.One representative model with the minimum error is considered as the best model,and we use the ensemble members near the best model in the cluster plane for applying EnKF.We demonstrate the proposed scheme for two 3D models that EnKF provides reliable assimilation results with much reduced computation time.

WEB SERVICE SELECTION ALGORITHM BASED ON PRINCIPAL COMPONENT ANALYSIS

Existing Web service selection approaches usually assume that preferences of users have been provided in a quantitative form by users. However, due to the subjectivity and vagueness of preferences, it may be impractical for users to specify quantitative and exact preferences. Moreover, due to that Quality of Service (QoS) attributes are often interrelated, existing Web service selection approaches which employ weighted summation of QoS attribute values to compute the overall QoS of Web services may produce inaccurate results, since they do not take correlations among QoS attributes into account. To resolve these problems, a Web service selection framework considering user's preference priority is proposed, which incorporates a searching mechanism with QoS range setting to identify services satisfying the user's QoS constraints. With the identified service candidates, based on the idea of Principal Component Analysis (PCA), an algorithm of Web service selection named PCA-WSS (Web Service Selection based on PCA) is proposed, which can eliminate the correlations among QoS attributes and compute the overall QoS of Web services accurately. After computing the overall QoS for each service, the algorithm ranks the Web service candidates based on their overall QoS and recommends services with top QoS values to users. Finally, the effectiveness and feasibility of our approach are validated by experiments, i.e. the selected Web service by our approach is given high average evaluation than other ones by users and the time cost of PCA-WSS algorithm is not affected acutely by the number of service candidates.

Grey Relational Analysis Coupled with Principal Component Analysis Method For Optimization Design of Novel Crash Box Structure

Crashworthiness and lightweight optimization design of the crash box are studied in this paper. For the initial model, a physical test was performed to verify the model. Then, a parametric model using mesh morphing technology is used to optimize and decrease the maximum collision force (MCF) and increase specific energy absorption (SEA) while ensure mass is not increased. Because MCF and SEA are two conflicting objectives, grey relational analysis (GRA) and principal component analysis (PCA) are employed for design optimization of the crash box. Furthermore, multi-objective analysis can convert to a single objective using the grey relational grade (GRG) simultaneously, hence, the proposed method can obtain the optimal combination of design parameters for the crash box. It can be concluded that the proposed method decreases the MCF and weight to 16.7% and 29.4% respectively, while increasing SEA to 16.4%. Meanwhile, the proposed method in comparison to the conventional NSGA-Ⅱ method, reduces the time cost by 103%. Hence, the proposed method can be properly applied to the optimization of the crash box.

Independent component analysis approach for fault diagnosis of condenser system in thermal power plant

A statistical signal processing technique was proposed and verified as independent component analysis(ICA) for fault detection and diagnosis of industrial systems without exact and detailed model.Actually,the aim is to utilize system as a black box.The system studied is condenser system of one of MAPNA's power plants.At first,principal component analysis(PCA) approach was applied to reduce the dimensionality of the real acquired data set and to identify the essential and useful ones.Then,the fault sources were diagnosed by ICA technique.The results show that ICA approach is valid and effective for faults detection and diagnosis even in noisy states,and it can distinguish main factors of abnormality among many diverse parts of a power plant's condenser system.This selectivity problem is left unsolved in many plants,because the main factors often become unnoticed by fault expansion through other parts of the plants.

Anomaly Detection System Based on Principal Component Analysis and Support Vector Machine

This article presents an anomaly detection system based on principal component analysis （PCA） and support vector machine （SVM）. The system first creates a profile defining a normal behavior by frequency-based scheme, and then compares the similarity of a current behavior with the created profile to decide whether the input instance is norreal or anomaly. In order to avoid overfitting and reduce the computational burden, normal behavior principal features are extracted by the PCA method. SVM is used to distinguish normal or anomaly for user behavior after training procedure has been completed by learning. In the experiments for performance evaluation the system achieved a correct detection rate equal to 92.2% and a false detection rate equal to 2.8%.

FUZZY PRINCIPAL COMPONENT ANALYSIS AND ITS KERNEL-BASED MODEL

Principal Component Analysis(PCA)is one of the most important feature extraction methods,and Kernel Principal Component Analysis(KPCA)is a nonlinear extension of PCA based on kernel methods.In real world,each input data may not be fully assigned to one class and it may partially belong to other classes.Based on the theory of fuzzy sets,this paper presents Fuzzy Principal Component Analysis(FPCA)and its nonlinear extension model,i.e.,Kernel-based Fuzzy Principal Component Analysis(KFPCA).The experimental results indicate that the proposed algorithms have good performances.

Effect of Two Kinds of Similarity Factors on Principal Component Analysis Fault Detection in Air Conditioning Systems

Screening similar historical fault-free candidate data would greatly affect the effectiveness of fault detection results based on principal component analysis(PCA).In order to find out the candidate data,this study compares unweighted and weighted similarity factors(SFs),which measure the similarity of the principal component subspace corresponding to the first k main components of two datasets.The fault detection employs the principal component subspace corresponding to the current measured data and the historical fault-free data.From the historical fault-free database,the load parameters are employed to locate the candidate data similar to the current operating data.Fault detection method for air conditioning systems is based on principal component.The results show that the weighted principal component SF can improve the effects of the fault-free detection and the fault detection.Compared with the unweighted SF,the average fault-free detection rate of the weighted SF is 17.33%higher than that of the unweighted,and the average fault detection rate is 7.51%higher than unweighted.

Relationship of public preferences and behavior in residential outdoor spaces using analytic hierarchy process and principal component analysis—a case study of Hangzhou City, China

This study examined public attitudes concerning the value of outdoor spaces which people use daily. Two successive analyses were performed based on data from common residents and college students in the city of Hangzhou, China. First, citizens registered various items constituting desirable values of residential outdoor spaces through a preliminary questionnaire. The result proposed three general attributes (functional, aesthetic and ecological) and ten specific qualities of residential outdoor spaces. An analytic hierarchy process (AHP) was applied to an interview survey in order to clarify the weights among these attributes and qualities. Second, principal factors were extracted from the ten specific qualities with principal component analysis (PCA) for both the common case and the campus case. In addition, the variations of respondents’ groups were classified with cluster analysis (CA) using the results of the PCA. The results of the AHP application found that the public prefers the functional attribute, rather than the aesthetic attribute. The latter is always viewed as the core value of open spaces in the eyes of architects and designers. Fur-thermore, comparisons of ten specific qualities showed that the public prefers the open spaces that can be utilized conveniently and easily for group activities, because such spaces keep an active lifestyle of neighborhood communication, which is also seen to protect human-regarding residential environments. Moreover, different groups of respondents diverge largely in terms of gender, age, behavior and preference.

Feature Extraction of Fabric Defects Based on Complex Contourlet Transform and Principal Component Analysis

To extract features of fabric defects effectively and reduce dimension of feature space,a feature extraction method of fabric defects based on complex contourlet transform(CCT)and principal component analysis(PCA)is proposed.Firstly,training samples of fabric defect images are decomposed by CCT.Secondly,PCA is applied in the obtained low-frequency component and part of highfrequency components to get a lower dimensional feature space.Finally,components of testing samples obtained by CCT are projected onto the feature space where different types of fabric defects are distinguished by the minimum Euclidean distance method.A large number of experimental results show that,compared with PCA,the method combining wavelet low-frequency component with PCA(WLPCA),the method combining contourlet transform with PCA(CPCA),and the method combining wavelet low-frequency and highfrequency components with PCA(WPCA),the proposed method can extract features of common fabric defect types effectively.The recognition rate is greatly improved while the dimension is reduced.

Influence of Three Sizes of Sliding Windows on Principle Component Analysis Fault Detection of Air Conditioning Systems

Principal component analysis(PCA)has been already employed for fault detection of air conditioning systems.The sliding window,which is composed of some parameters satisfying with thermal load balance,can select the target historical fault-free reference data as the template which is similar to the current snapshot data.The size of sliding window is usually given according to empirical values,while the influence of different sizes of sliding windows on fault detection of an air conditioning system is not further studied.The air conditioning system is a dynamic response process,and the operating parameters change with the change of the load,while the response of the controller is delayed.In a variable air volume(VAV)air conditioning system controlled by the total air volume method,in order to ensure sufficient response time,30 data points are selected first,and then their multiples are selected.Three different sizes of sliding windows with 30,60 and 90 data points are applied to compare the fault detection effect in this paper.The results show that if the size of the sliding window is 60 data points,the average fault-free detection ratio is 80.17%in fault-free testing days,and the average fault detection ratio is 88.47%in faulty testing days.

Computational Intelligence Prediction Model Integrating Empirical Mode Decomposition,Principal Component Analysis,and Weighted k-Nearest Neighbor

On the basis of machine leaning,suitable algorithms can make advanced time series analysis.This paper proposes a complex k-nearest neighbor(KNN)model for predicting financial time series.This model uses a complex feature extraction process integrating a forward rolling empirical mode decomposition(EMD)for financial time series signal analysis and principal component analysis(PCA)for the dimension reduction.The information-rich features are extracted then input to a weighted KNN classifier where the features are weighted with PCA loading.Finally,prediction is generated via regression on the selected nearest neighbors.The structure of the model as a whole is original.The test results on real historical data sets confirm the effectiveness of the models for predicting the Chinese stock index,an individual stock,and the EUR/USD exchange rate.

Sparse flight spotlight mode 3-D imaging of spaceborne SAR based on sparse spectrum and principal component analysis

The spaceborne synthetic aperture radar(SAR)sparse flight 3-D imaging technology through multiple observations of the cross-track direction is designed to form the cross-track equivalent aperture,and achieve the third dimensionality recognition.In this paper,combined with the actual triple star orbits,a sparse flight spaceborne SAR 3-D imaging method based on the sparse spectrum of interferometry and the principal component analysis(PCA)is presented.Firstly,interferometric processing is utilized to reach an effective sparse representation of radar images in the frequency domain.Secondly,as a method with simple principle and fast calculation,the PCA is introduced to extract the main features of the image spectrum according to its principal characteristics.Finally,the 3-D image can be obtained by inverse transformation of the reconstructed spectrum by the PCA.The simulation results of 4.84 km equivalent cross-track aperture and corresponding 1.78 m cross-track resolution verify the effective suppression of this method on high-frequency sidelobe noise introduced by sparse flight with a sparsity of 49%and random noise introduced by the receiver.Meanwhile,due to the influence of orbit distribution of the actual triple star orbits,the simulation results of the sparse flight with the 7-bit Barker code orbits are given as a comparison and reference to illuminate the significance of orbit distribution for this reconstruction results.This method has prospects for sparse flight 3-D imaging in high latitude areas for its short revisit period.