Uncertainty handling using neural network-based prediction intervals for electrical load forecasting

The complexity and level of uncertainty present in operation of power systems have significantly grown due to penetration of renewable resources. These complexities warrant the need for advanced methods for load forecasting and quantifying uncertainties associated with forecasts. The objective of this study is to develop a framework for probabilistic forecasting of electricity load demands. The proposed probabilistic framework allows the analyst to construct PIs (prediction intervals) for uncertainty quantification. A newly introduced method, called LUBE (lower upper bound estimation), is applied and extended to develop PIs using NN (neural network) models. The primary problem for construction of intervals is firstly formulated as a constrained single-objective problem. The sharpness of PIs is treated as the key objective and their calibration is considered as the constraint. PSO (particle swarm optimization) enhanced by the mutation operator is then used to optimally tune NN parameters subject to constraints set on the quality of PIs. Historical load datasets from Singapore, Ottawa (Canada) and Texas (USA) are used to examine performance of the proposed PSO-based LUBE method. According to obtained results, the proposed probabilistic forecasting method generates well-calibrated and informative PIs. Furthermore, comparative results demonstrate that the proposed PI construction method greatly outperforms three widely used benchmark methods. © 2014 Elsevier Ltd.

Neural network training by hybrid accelerated cuckoo particle swarm optimization algorithm

Metaheuristic algorithm is one of the most popular methods in solving many optimization problems. This paper presents a new hybrid approach comprising of two natures inspired metaheuristic algorithms i.e. Cuckoo Search (CS) and Accelerated Particle Swarm Optimization (APSO) for training Artificial Neural Networks (ANN). In order to increase the probability of the egg’s survival, the cuckoo bird migrates by traversing more search space. It can successfully search better solutions by performing levy flight with APSO. In the proposed Hybrid Accelerated Cuckoo Particle Swarm Optimization (HACPSO) algorithm, the communication ability for the cuckoo birds have been provided by APSO, thus making cuckoo bird capable of searching for the best nest with better solution. Experimental results are carried-out on benchmarked datasets, and the performance of the proposed hybrid algorithm is compared with Artificial Bee Colony (ABC) and similar hybrid variants. The results show that the proposed HACPSO algorithm performs better than other algorithms in terms of convergence and accuracy.

Application of neural network to rock slope stability assessments

It is known that rock masses are inhomogeneous, discontinuous media composed of rock material and naturally occurring discontinuities such as joints, fractures and bedding planes. These features make any analysis very difficult using simple theoretical solutions. Generally speaking, back analysis technique can be used to capture some implicit parameters for geotechnical problems. In order to perform back analyses, the procedure of trial and error is generally required. However, it would be time-consuming. This study aims at applying a neural network to do the back analysis for rock slope failures. The neural network tool will be trained by using the solutions of finite element upper and lower bound limit analysis methods. Therefore, the uncertain parameter can be obtained, particularly for rock mass disturbance.

Spiking neural network-based target tracking control for autonomous mobile robots

In this paper, a target tracking controller based on spiking neural network is proposed for autonomous robots. This controller encodes the preprocessed environmental and target information provided by CCD cameras, encoders and ultrasonic sensors into spike trains, which are integrated by a three-layer spiking neural network (SNN). The outputs of SNN are generated based on the competition between the forward/backward neuron pair corresponding to each motor, with the weights evolved by the Hebbian learning. The application to target tracking of a mobile robot in unknown environment verifies the validity of the proposed controller.

An incremental meta-cognitive-based scaffolding fuzzy neural network

The idea of meta-cognitive learning has enriched the landscape of evolving systems, because it emulates three fundamental aspects of human learning: what-to-learn; how-to-learn; and when-to-learn. However, existing meta-cognitive algorithms still exclude Scaffolding theory, which can realize a plug-and-play classifier. Consequently, these algorithms require laborious pre- and/or post-training processes to be carried out in addition to the main training process. This paper introduces a novel meta-cognitive algorithm termed GENERIC-Classifier (gClass), where the how-to-learn part constitutes a synergy of Scaffolding Theory - a tutoring theory that fosters the ability to sort out complex learning tasks, and Schema Theory - a learning theory of knowledge acquisition by humans. The what-to-learn aspect adopts an online active learning concept by virtue of an extended conflict and ignorance method, making gClass an incremental semi-supervised classifier, whereas the when-to-learn component makes use of the standard sample reserved strategy. A generalized version of the Takagi-Sugeno Kang (TSK) fuzzy system is devised to serve as the cognitive constituent. That is, the rule premise is underpinned by multivariate Gaussian functions, while the rule consequent employs a subset of the non-linear Chebyshev polynomial. Thorough empirical studies, confirmed by their corresponding statistical tests, have numerically validated the efficacy of gClass, which delivers better classification rates than state-of-the-art classifiers while having less complexity.

A new genetic algorithm based on modified Physarum network model for bandwidth-delay constrained least-cost multicast routing

A mobile ad hoc network is a kind of popular self-configuring network, in which multicast routing under the quality of service constraints, is a significant challenge. Many researchers have proved that such problem can be formulated as a NP-complete problem and proposed some swarm-based intelligent algorithms to solve the optimal solution, such as the genetic algorithm (GA), bees algorithm. However, a lower efficiency of local search ability and weak robustness still limit the computational effectiveness. Aiming to those shortcomings, a new hybrid algorithm inspired by the self-organization of Physarum, is proposed in this paper. In our algorithm, an updating scheme based on Physarum network model (PM) is used for improving the crossover operator of traditional GAs, in which the same parts of parent chromosomes are reserved and the new offspring by the PM is generated. In order to estimate the effectiveness of our proposed optimized scheme, some typical genetic algorithms and their updating algorithms (PMGAs) are compared for solving the multicast routing on four different datasets. The simulation experiments show that PMGAs are more efficient than original GAs. More importantly, the PMGAs are more robustness that is very important for solving the multicast routing problem. Moreover, a series of parameter analyses is used to find a set of better setting for realizing the maximal efficiency of our algorithm.

Rock slope stability analyses using extreme learning neural network and terminal steepest descent algorithm

The analysis of rock slope stability is a classical problem for geotechnical engineers. However, for practicing engineers, proper software is not usually user friendly, and additional resources capable of providing information useful for decision-making are required. This study developed a convenient tool that can provide a prompt assessment of rock slope stability. A nonlinear input-output mapping of the rock slope system was constructed using a neural network trained by an extreme learning algorithm. The training data was obtained by using finite element upper and lower bound limit analysis methods. The newly developed techniques in this study can either estimate the factor of safety for a rock slope or obtain the implicit parameters through back analyses. Back analysis parameter identification was performed using a terminal steepest descent algorithm based on the finite-time stability theory. This algorithm not only guarantees finite-time error convergence but also achieves exact zero convergence, unlike the conventional steepest descent algorithm in which the training error never reaches zero.

The prediction of damage condition in regards to damage factor influence of light structures on expansive soils in Victoria, Australia

This paper proposes a neural network model using genetic algorithm for a model for the prediction of the damage condition of existing light structures founded in expansive soils in Victoria, Australia. It also accounts for both individual effects and interactive effects of the damage factors influencing the deterioration of light structures. A Neural Network Model was chosen because it can deal with 'noisy' data while a Genetic Algorithm was chosen because it does not get `trapped' in local optimum like other gradient descent methods. The results obtained were promising and indicate that a Neural Network Model trained using a Genetic Algorithm has the ability to develop an interactive relationship and a Predicted Damage Conditions Model.

Integrating simulated annealing and delta technique for constructing optimal prediction intervals

This paper aims at developing a new criterion for quantitative assessment of prediction intervals. The proposed criterion is developed based on both key measures related to quality of prediction intervals: length and coverage probability. This criterion is applied as a cost function for optimizing prediction intervals constructed using delta technique for neural network model. Optimization seeks out to minimize length of prediction intervals without compromising their coverage probability. Simulated Annealing method is employed for readjusting neural network parameters for minimization of the new cost function. To further ameliorate search efficiency of the optimization method, parameters of the network trained using weight decay method are considered as the initial set in Simulated Annealing algorithm. Implementation of the proposed method for a real world case study shows length and coverage probability of constructed prediction intervals are better than those constructed using traditional techniques.

The ranking of factors influencing the behaviour of light structures on expansive soils in Victoria, Australia

In this paper a Neural Network Model was used to develop a ranking of the potential damage influences for light structures on expansive soils in Victoria. These influences include geology, Thornthwaite moisture index, vegetation covers, construction foundation type, construction wall type, geographical region and age of building when first inspected. Approximately 400 cases of damage to light structures in Victoria, Australia were considered in this study. Feedforward Backpropagation was adopted to train the data. The ranking of importance was estimated using connection weight approach and then compared to results calculated from sensitivity analysis. From the analysis, the ranking of importance for potential damage factor was noted.

Intelligent traffic light control of isolated intersections using machine learning methods

Traffic congestion is one of the major problems in modern cities. This study applies machine learning methods to determine green times in order to minimize in an isolated intersection. Q-learning and neural networks are applied here to set signal light times and minimize total delays. It is assumed that an intersection behaves in a similar fashion to an intelligent agent learning how to set green times in each cycle based on traffic information. Here, a comparison between Q-learning and neural network is presented. In Q-learning, considering continuous green time requires a large state space, making the learning process practically impossible. In contrast to Q-learning methods, the neural network model can easily set the appropriate green time to fit the traffic demand. The performance of the proposed neural network is compared with two traditional alternatives for controlling traffic lights. Simulation results indicate that the application of the proposed method greatly reduces the total delay in the network compared to the alternative methods.

Damage identification scheme based on compressive sensing

Civil infrastructures are critical to every nation, due to their substantial investment, long service period, and enormous negative impacts after failure. However, they inevitably deteriorate during their service lives. Therefore, methods capable of assessing conditions and identifying damage in a structure timely and accurately have drawn increasing attention. Recently, compressive sensing (CS), a significant breakthrough in signal processing, has been proposed to capture and represent compressible signals at a rate significantly below the traditional Nyquist rate. Due to its sound theoretical background and notable influence, this methodology has been successfully applied in many research areas. In order to explore its application in structural damage identification, a new CS-based damage identification scheme is proposed in this paper, by regarding damage identification problems as pattern classification problems. The time domain structural responses are transferred to the frequency domain as sparse representation, and then the numerical simulated data under various damage scenarios will be used to train a feature matrix as input information. This matrix can be used for damage identification through an optimization process. This will be one of the first few applications of this advanced technique to structural engineering areas. In order to demonstrate its effectiveness, numerical simulation results on a complex pipe soil interaction model are used to train the parameters and then to identify the simulated pipe degradation damage and free-spanning damage. To further demonstrate the method, vibration tests of a steel pipe laid on the ground are carried out. The measured acceleration time histories are used for damage identification. Both numerical and experimental verification results confirm that the proposed damage identification scheme will be a promising tool for structural health monitoring.

New artificial intelligence based tire size identification for fast and safe inflating cycle

Motor vehicle accidents are one of the main killers on the road. Modern vehicles have several safety features to improve the stability and controllability. The tire condition is critical to the proper function of the designed safety features. Under or over inflated tires adversely affects the stability of vehicles. It is generally the vehicle's user responsibility to ensure the tire inflation pressure is set and maintained to the required value using a tire inflator. In the tire inflator operation, the vehicle's user sets the desired value and the machine has to complete the task. During the inflation process, the pressure sensor does not read instantaneous static pressure to ensure the target value is reached. Hence, the inflator is designed to stop repetitively for pressure reading and avoid over inflation. This makes the inflation process slow, especially for large tires. This paper presents a novel approach using artificial neural network based technique to identify the tire size. Once the tire size is correctly identified, an optimized inflation cycle can be computed to improve performance, speed and accuracy of the inflation process. The developed neural network model was successfully simulated and tested for predicting tire size from the given sets of input parameters. The test results are analyzed and discussed in this paper.

Improving the quality of load forecasts using smart meter data

For the operator of a power system, having an accurate forecast of the day-ahead load is imperative in order to guaranty the reliability of supply and also to minimize generation costs and pollution. Furthermore, in a restructured power system, other parties, like utility companies, large consumers and in some cases even ordinary consumers, can benefit from a higher quality demand forecast. In this paper, the application of smart meter data for producing more accurate load forecasts has been discussed. First an ordinary neural network model is used to generate a forecast for the total load of a number of consumers. The results of this step are used as a benchmark for comparison with the forecast results of a more sophisticated method. In this new method, using wavelet decomposition and a clustering technique called interactive k-means, the consumers are divided into a number of clusters. Then for each cluster an individual neural network is trained. Consequently, by adding the outputs of all of the neural networks, a forecast for the total load is generated. A comparison between the forecast using a single model and the forecast generated by the proposed method, proves that smart meter data can be used to significantly improve the quality of load forecast.