Short-term solar energy forecasting: Integrated computational intelligence of LSTMs and GRU.

Problems with erroneous forecasts of electricity production from solar farms create serious operational, technological, and financial challenges to both Solar farm owners and electricity companies. Accurate prediction results are necessary for efficient spinning reserve planning as well as regulating inertia and power supply during contingency events. In this work, the impact of several climatic conditions on solar electricity generation in Amherst. Furthermore, three machine learning models using Lasso Regression, ridge Regression, ElasticNet regression, and Support Vector Regression, as well as deep learning models for time series analysis include long short-term memory, bidirectional LSTM, and gated recurrent unit along with their variants for estimating solar energy generation for every five-minute interval on Amherst weather power station. These models were evaluated using mean absolute error root means square error, mean square error, and mean absolute percentage error. It was observed that horizontal solar irradiance and water saturation deficiency had a highly proportional relationship with Solar PV electricity generation. All proposed machine learning models turned out to perform well in predicting electricity generation from the analyzed solar farm. Bi-LSTM has performed the best among all models with 0.0135, 0.0315, 0.0012, and 0.1205 values of MAE, RMSE, MSE, and MAPE, respectively. Comparison with the existing methods endorses the use of our proposed RNN variants for higher efficiency, accuracy, and robustness. Multistep-ahead solar energy prediction is also carried out by exploiting hybrids of LSTM, Bi-LSTM, and GRU.

PCGA: Polynomial collocation genetic algorithm for singular Poisson-Boltzmann equation arising in thermal explosions.

Heat generation as a result of the exothermic reaction reaches the environment mainly due to the conduction through the walls of the vessel. The balance between the heat generated and the heat conducted away, resulting in the explosion is described by the Frank-Kamenetzkii (FK) parameter ρ. The critical value of FK for which the explosion occurs depends upon the shape of the vessel, which requires the solution of governing singular nonlinear Poisson-Boltzmann equation. Owing to the exponential nonlinearity and singularity the analytical exact solution for the non-integer k values does not exist. This work focuses on implementing the polynomial collocation by exploiting the global optimization features of the genetic algorithm to solve the Poisson-Boltzmann equation for integer and non-integer shape factors (k). The governing equation was converted into coupled nonlinear algebraic equations and an objective function was formulated. The method was examined for six different configurations of the control parameters of GA to find the best set of parameters. The solution for temperature distribution is obtained for cylindrical (k = 1), parallelepiped (k = 0.438, 0.694), and an arbitrary shape (k = 0.5) respectively. The solution obtained from Polynomial Collocation Genetic Algorithm (PCGA) remained in good agreement with the corresponding analytical results for k = 1, with the minimum absolute error of 10 - 10 . The critical values of the FK are obtained as 1.5 , 1.4 , a n d 1.7 for shape factor k = 0.438 , 0.5 , a n d 0.694 respectively with the convergence of the order of 10 - 6 t o 10 - 5 . The obtained solution is fairly stable over appropriate independent runs with the variation in the fitness value ranging from 10 - 05 t o 10 - 03 . Further simulations were performed to validate the results through statistical error indices. The diminutive errors of the order of 10 - 6 confirm reliable optimum solution, accuracy, and stability.

A simple, sensitive, label-free electrochemical immunosensor based on the chitosan-coated silver/cerium oxide (CS@Ag/CeO2) nanocomposites for the detection of alpha-fetoprotein (AFP).

Metal oxide-based sensors have the benefit of inexpensive, quick response, and high sensitivity in detecting specific biological species. In this article, a simple electrochemical immunosensor was fabricated using antibody-chitosan coated silver/cerium oxide (Ab-CS@Ag/CeO2) nanocomposites on a gold electrode for sensitive alpha-fetoprotein (AFP) diagnosis in human serum samples. Successfully synthesis of AFP antibody-CS@Ag/CeO2conjugates was confirmed through Fourier transform infrared spectra of the prototype. The amine coupling bond chemistry was then used to immobilize the resultant conjugate on a gold electrode surface. It was observed that the interaction of the synthesized Ab-CS@Ag/CeO2nanocomposites with AFP prevented an electron transfer and reduced the voltammetric Fe(CN)63-/4-peak current, which was proportional to the amount of AFP. The linear ranges of AFP concentration were found from 10-12-10-6g.ml-1. The limit of detection was calculated using the calibration curve and came out to be 0.57 pg.ml-1. The designed label-free immunosensor successfully detected AFP in human serum samples. As a result, the resulting immunosensor is a promising sensor plate form for AFP detection and could be used in clinical bioanalysis.

Reusable, Noninvasive, and Sensitive Fluorescence Enhanced ZnO-Nanorod-Based Microarrays for Quantitative Detection of AFP in Human Serum.

The fabrication of sensitive protein microarrays such as PCR used in DNA microarray is challenging due to lack of signal amplification. The development of microarrays is utilized to improve the sensitivity and limitations of detection towards primal cancer detection. The sensitivity is enhanced by the use of ZnO-nanorods and is investigated as a substrate which enhance the florescent signal to diagnose the hepatocellular carcinoma (HCC) at early stages. The substrate for deposition of ZnO-nanorods is prepared by the conventional chemical bath deposition method. The resultant highly dense ZnO-nanorods enhance the fluorescent signal 7.2 times as compared to the substrate without ZnO-nanorods. The microarray showed sensitivity of 1504.7 ng ml-1 and limit of detection of 0.1 pg ml-1 in wide dynamic range of 0.05 pg-10 µg ml-1 for alpha fetoprotein (AFP) detection in 10% human serum. This immunoassay was successfully applied for human serum samples to detect tumor marker with good recoveries. The ZnO-nanorod substrate is a simple protein microarray which showed a great promise for developing a low-cost, sensitive, and high-throughput protein assay platform for several applications in both fundamental research and clinical diagnosis.