Fuzzy Clustering-Based Deep Learning for Short-Term Load Forecasting in Power Grid Systems Using Time-Varying and Time-Invariant Features.

Accurate short-term load forecasting (STLF) is essential for power grid systems to ensure reliability, security and cost efficiency. Thanks to advanced smart sensor technologies, time-series data related to power load can be captured for STLF. Recent research shows that deep neural networks (DNNs) are capable of achieving accurate STLP since they are effective in predicting nonlinear and complicated time-series data. To perform STLP, existing DNNs use time-varying dynamics of either past load consumption or past power correlated features such as weather, meteorology or date. However, the existing DNN approaches do not use the time-invariant features of users, such as building spaces, ages, isolation material, number of building floors or building purposes, to enhance STLF. In fact, those time-invariant features are correlated to user load consumption. Integrating time-invariant features enhances STLF. In this paper, a fuzzy clustering-based DNN is proposed by using both time-varying and time-invariant features to perform STLF. The fuzzy clustering first groups users with similar time-invariant behaviours. DNN models are then developed using past time-varying features. Since the time-invariant features have already been learned by the fuzzy clustering, the DNN model does not need to learn the time-invariant features; therefore, a simpler DNN model can be generated. In addition, the DNN model only learns the time-varying features of users in the same cluster; a more effective learning can be performed by the DNN and more accurate predictions can be achieved. The performance of the proposed fuzzy clustering-based DNN is evaluated by performing STLF, where both time-varying features and time-invariant features are included. Experimental results show that the proposed fuzzy clustering-based DNN outperforms the commonly used long short-term memory networks and convolution neural networks.

Improving Text-Independent Forced Alignment to Support Speech-Language Pathologists with Phonetic Transcription.

Problem: Phonetic transcription is crucial in diagnosing speech sound disorders (SSDs) but is susceptible to transcriber experience and perceptual bias. Current forced alignment (FA) tools, which annotate audio files to determine spoken content and its placement, often require manual transcription, limiting their effectiveness. Method: We introduce a novel, text-independent forced alignment model that autonomously recognises individual phonemes and their boundaries, addressing these limitations. Our approach leverages an advanced, pre-trained wav2vec 2.0 model to segment speech into tokens and recognise them automatically. To accurately identify phoneme boundaries, we utilise an unsupervised segmentation tool, UnsupSeg. Labelling of segments employs nearest-neighbour classification with wav2vec 2.0 labels, before connectionist temporal classification (CTC) collapse, determining class labels based on maximum overlap. Additional post-processing, including overfitting cleaning and voice activity detection, is implemented to enhance segmentation. Results: We benchmarked our model against existing methods using the TIMIT dataset for normal speakers and, for the first time, evaluated its performance on the TORGO dataset containing SSD speakers. Our model demonstrated competitive performance, achieving a harmonic mean score of 76.88% on TIMIT and 70.31% on TORGO. Implications: This research presents a significant advancement in the assessment and diagnosis of SSDs, offering a more objective and less biased approach than traditional methods. Our model's effectiveness, particularly with SSD speakers, opens new avenues for research and clinical application in speech pathology.

Developing Nonlinear Customer Preferences Models for Product Design Using Opining Mining and Multiobjective PSO-Based ANFIS Approach.

Online customer reviews can clearly show the customer experience, and the improvement suggestions based on the experience, which are helpful to product optimization and design. However, the research on establishing a customer preference model based on online customer reviews is not ideal, and the following research problems are found in previous studies. Firstly, the product attribute is not involved in the modelling if the corresponding setting cannot be found in the product description. Secondly, the fuzziness of customers' emotions in online reviews and nonlinearity in the models were not appropriately considered. Thirdly, the adaptive neuro-fuzzy inference system (ANFIS) is an effective way to model customer preferences. However, if the number of inputs is large, the modelling process will be failed due to the complex structure and long computational time. To solve the above-given problems, this paper proposed multiobjective particle swarm optimization (PSO) based ANFIS and opinion mining, to build customer preference model by analyzing the content of online customer reviews. In the process of online review analysis, the opinion mining technology is used to conduct comprehensive analysis on customer preference and product information. According to the analysis of information, a new method for establishing customer preference model is proposed, that is, a multiobjective PSO based ANFIS. The results show that the introducing of multiobjective PSO method into ANFIS can effectively solve the defects of ANFIS itself. Taking hair dryer as a case study, it is found that the proposed approach performs better than fuzzy regression, fuzzy least-squares regression, and genetic programming based fuzzy regression in modelling customer preference.

Developing explicit customer preference models using fuzzy regression with nonlinear structure.

In online sales platforms, product design attributes influence consumer preferences, and consumer preferences also have a significant impact on future product design optimization and iteration. Online review data are the most intuitive feedback from consumers on products. Using the value of online review information to explore consumer preferences is the key to optimize the products, improve consumer satisfaction and meet consumer requirements. Therefore, the study of consumer preferences based on online reviews is of great importance. However, in previous research on consumer preferences based on online reviews, few studies have modeled consumer preferences. The models often suffer from the nonlinear structure and the fuzzy coefficients, making it challenging to build explicit models. Therefore, this study adopts a fuzzy regression approach with a nonlinear structure to model consumer preferences based on online reviews to provide reference and insight for subsequent studies. First, smartwatches were selected as the research object, and the sentiment scores of product reviews under different topics were obtained by text mining on the product online data. Second, a polynomial structure between product attributes and consumer preferences was generated to investigate the association between them further. Afterward, based on the existing polynomial structure, the fuzzy coefficients of each item in the structure were determined by the fuzzy regression approach. Finally, the mean relative error and mean systematic confidence of the fuzzy regression with nonlinear structure method were numerically calculated and compared with fuzzy least squares regression, fuzzy regression, adaptive neuro fuzzy inference system (ANFIS) and K-means-based ANFIS, and it was found that the proposed method was relatively more effective in modeling consumer preferences.

Pulsatile blood flow and gas exchange across a cylindrical fiber array.

The pulsatile blood flow and gas transport of oxygen and carbon dioxide through a cylindrical array of microfibers are numerically simulated. Blood is modeled as a homogeneous Casson fluid, and hemoglobin molecules in blood are assumed to be in local equilibrium with oxygen and carbon dioxide. It is shown that flow pulsatility enhances gas transport and the amount of gas exchange is sensitive to the blood flow field across the fibers. The steady Sherwood number dependence on Reynolds number was shown to have a linear relation consistent with experimental findings. For most cases, an enhancement in gas transport is accompanied with an increase in flow resistance. Maximum local shear stress is provided as a possible indicator of thrombosis, and the computed shear stress is shown to be below the threshold value for thrombosis formation for all cases evaluated.

Shear modulation of intercellular contact area between two deformable cells colliding under flow.

Shear rate has been shown to critically affect the kinetics and receptor specificity of cell-cell interactions. In this study, the collision process between two modeled cells interacting in a linear shear flow is numerically investigated. The two identical biological or artificial cells are modeled as deformable capsules composed of an elastic membrane. The cell deformation and trajectories are computed using the immersed boundary method (IBM) for shear rates of 100-400s(-1). As the two cells collide under hydrodynamic shear, large local cell deformations develop. The effective contact area between the two cells is modulated by the shear rate, and reaches a maximum value at intermediate levels of shear. At relatively low shear rate, the contact area is an enclosed region. As the shear rate increases, dimples form on the membrane surface, and the contact region becomes annular. The nonmonotonic increase of the contact area with the increase of shear rate from computational results implies that there is a maximum effective receptor-ligand binding area for cell adhesion. This finding suggests the existence of possible hydrodynamic mechanism that could be used to interpret the observed maximum leukocyte aggregation in shear flow. The critical shear rate for maximum intercellular contact area is shown to vary with cell properties such as radius and membrane elastic modulus.

Spontaneous spreading of surfactant-bearing drops in the sorption-controlled limit.

Axisymmetric spreading of a liquid drop containing a soluble surfactant on a smooth solid substrate is numerically investigated for the case in which surfactant mass transfer between the interface and the bulk liquid is sorption/kinetic controlled. The fastest spreading rate is achieved by drops with O(1) values of Biot number for which the rate of surface convection is comparable to the sorption rate, and the surfactant molecules transferred to the interface are effectively convected to the contact line region.

Pulsatile flow and mass transport over an array of cylinders: gas transfer in a cardiac-driven artificial lung.

The pulsatile flow and gas transport of a Newtonian passive fluid across an array of cylindrical microfibers are numerically investigated. It is related to an implantable, artificial lung where the blood flow is driven by the right heart. The fibers are modeled as either squared or staggered arrays. The pulsatile flow inputs considered in this study are a steady flow with a sinusoidal perturbation and a cardiac flow. The aims of this study are twofold: identifying favorable array geometry/spacing and system conditions that enhance gas transport; and providing pressure drop data that indicate the degree of flow resistance or the demand on the right heart in driving the flow through the fiber bundle. The results show that pulsatile flow improves the gas transfer to the fluid compared to steady flow. The degree of enhancement is found to be significant when the oscillation frequency is large, when the void fraction of the fiber bundle is decreased, and when the Reynolds number is increased; the use of a cardiac flow input can also improve gas transfer. In terms of array geometry, the staggered array gives both a better gas transfer per fiber (for relatively large void fraction) and a smaller pressure drop (for all cases). For most cases shown, an increase in gas transfer is accompanied by a higher pressure drop required to power the flow through the device.

Surfactant-assisted spreading of a liquid drop on a smooth solid surface.

Axisymmetric spreading of a liquid drop covered with an insoluble surfactant monolayer on a smooth solid substrate is numerically investigated. As the drop spreads, the adsorbed surfactant molecules are constantly redistributed along the air-liquid interface by convection and diffusion, leading to nonuniformities in surface tension along the interface. The resulting Marangoni stresses affect the spreading rate by altering the surface flow and the drop shape. In addition, surfactant accumulation in the vicinity of the moving contact line affects the spreading rate by altering the balance of line forces. Two different models for the constitutive relation at the moving contact line are used, in conjunction with a surface equation of state based on the Frumkin adsorption framework, to probe the surfactant influence. The coupled evolution equations for the drop shape and monolayer concentration profile are integrated using a pseudospectral method to determine the rate of surfactant-assisted spreading over a wide range of the dimensionless parameters governing the spreading process. The insoluble monolayer enhances spreading through two mechanisms; a reduction in the equilibrium contact angle, and an increase in the magnitude of the radial pressure gradient within the drop due to the formation of positive surface curvature near the moving contact line. Both mechanisms are driven by the accumulation of surfactant at the contact line due to surface convection. Although the Marangoni stresses induced at the air-liquid interface reduce the rate of spreading during the initial stages of spreading, their retarding effect is overwhelmed by the favorable effects of the aforementioned mechanisms to lead to an overall enhancement in the rate of spreading in most cases. The spreading rate is found to be higher for bulkier surfactants with stronger repulsive interactions. With the exception of monolayers with strong cohesive interactions which tend to retard the spreading process, the overall effect of an insoluble monolayer is to increase the rate of drop spreading. Simulation results for small Bond numbers indicate the existence of a power-law region for the time-dependence of the basal radius of the drop, consistent with experimental measurements.