International Journal of Computational Intelligence and Applications

In this paper, two intelligent strategies for energy management unit for a home integrated with smart grid are proposed. The strategies are based on classical Boolean and genetic algorithm (GA). The objective is to optimize the cost saving for the end consumer. The price of energy varies by the hour depending on the load on the grid. The two strategies predict when and by how much the storage unit installed in the house should charge and release for 24 h of the day, satisfying the constraint that the load demand of the house at any particular hour should always be met. The strategies were tested by real time data collected by the Department of Energy for a typical house in the Chicago, Illinois region for the year 2013. Both the strategies achieve cost savings; however, it has been found that GA-based strategy results in higher cost saving. The impact of the capacity of the energy storage unit (ESU) on the cost saving has been analyzed for a GA strategy and cost saving obtained when the capacity of ESU is 1.5 times and 2 times the house hold load at any given hour is presented.

Although lots of ways can be used to estimate geo-stress state, estimation of geo-stress state without knowing geomechanical parameters such as pore pressure, tensile strength and Poisson’s ratio, etc., still remains one of the most challenging tasks in geotechnical engineering. The main contribution of this paper is to present a back-propagation neural network (BPNN) with genetic algorithm (GA) optimization to predict the geo-stresses based on wellbore pressures of hydraulic fracturing tests during drilling. In the suggested hybrid model, the BPNN is used establish a mapping between the recording pressures and the geo-stress state. Also the GA is used to carry out the optimization of the weights and thresholds of BPNN model for improving accuracy of prediction. Finally, based on the record pressures in hydraulic fracturing (HF) tests, the BPNN model with genetic algorithm optimization successfully predicts the geo-stresses at the corresponding formation in the event that these parameters such as pore pressure, tensile strength and Poisson’s ratio are unavailable. In the meantime, the geo-stress state has been calculated using the theoretical formula by assuming pore pressure and tensile strength of rock mass are known. Then results from theoretical equation, BPNN and BPNN with GA optimization are compared, which shows that the degree accuracy of geo-stresses predicted by using GA-BPNN model is more obviously improvement than the predicted results by the basic BPNN model.

This paper introduces a dynamic multi-objective optimization algorithm integrated of Cat swarm algorithm and Borda count ranking method. In the proposed approach, the cats of population are ranked and sorted based on Borda selection method just before and after a change. Then the cats with the worst Borda’s rank are re-initialized to improve the diversity of population. Also, a multi-objective cat swarm optimization (CSO) as a population-based evolutionary algorithm is applied to converge to optimal front. The simulation results indicate that the proposed algorithm achieves competitive results in comparison with traditional approaches.

Rainfall-runoff modeling is very important for Water Resources Management because accurate and timely prediction can avoid accidents, such as the life risk, economic losses, etc. This paper proposed the novel hybrid optimization algorithm to combine Neural Network (NN) for rainfall-runoff modeling, namely HGASA-NN. Firstly, a novel and specialized hybrid optimization strategy by incorporating Simulated Annealing algorithm (SA) into Genetic Algorithm (GA) was used to train the initial connection weights and thresholds of NN. Secondly, the Back Propagation (BP) algorithm was adjusted the final weights and biases. Finally, a numerical example of daily rainfall-runoff data was used to elucidate the forecasting performance of the proposed HGASA-NN model. The HGASA-NN can make use of not only strong global searching ability of the GASA, but also strong local searching ability of the BP algorithm. The forecasting results indicate that the proposed model yields more accurate forecasting results than the BP-NN and pure GA training NN model. Therefore, the HGASA-NN model is a promising alternative for rainfall-runoff forecasting.

Weight and bias connection are important features of neural networks, which is still challenging for researchers. In this work, we focus on initial weights and bias connection of counter propagation network (CPN) using modified chaotic bat algorithm (MCBA) i.e., MCBA-CPN for uncertain nonlinear systems and compare it with CPN using chaotic bat algorithm (CBA) i.e., CBA-CPN. Chaotic function is used for pulse frequency of bats in MCBA. We have implemented CBA and MCBA, which are based on the consideration of the global solution in the sound intensity adjustment. MCBA-CPN is applied on different uncertain nonlinear systems and Mackey–Glass time series data to test the concert in terms of prediction accuracy. Proposed method is validated through statistical testing like chi-square and $t$-test demonstrate that the difference between target and output of proposed method are acceptable. Finally, MCBA-CPN is applied to a real world problem for prediction of milk production data.

Despite some success of Genetic Algorithms (GAs) when tackling Constraint Satisfaction Problems (CSPs), they generally suffer from poor crossover operators. In order to overcome this limitation in practice, we propose a novel crossover specifically designed for solving CSPs including Temporal CSPs (TCSPs). Together with a variable ordering heuristic and an integration into a parallel architecture, this proposed crossover enables the solving of large and hard problem instances as demonstrated by the experimental tests conducted on randomly generated CSPs and TCSPs based on the model RB. We will indeed demonstrate, through these tests, that our proposed method is superior to the known GA-based techniques for CSPs. In addition, we will show that we are able to compete with the efficient MAC-based Abscon 109 solver for random problem instances as well as those instances taken from Lecoutre’s CSP library. Finally, we conducted additional tests on very large consistent and over constrained CSPs and TCSPs instances in order to show the ability of our method to deal with constraint problems in real time. This corresponds to solving the CSP or the TCSP by giving a solution with a quality (number of solved constraints) depending on the time allocated for computation.

This paper proposes a novel method which is based on particle swarm optimization (PSO) algorithm for the two-dimensional FIR digital filter design. In the PSO algorithm, it simply uses two adjusting mechanisms including particle’s velocity and position updating to achieve the optimization. In addition, numerical representations for each candidate solution are completely real numbers. The PSO algorithm is utilized to design the two-dimensional FIR digital filter with linear-phase characteristic so that its frequency response can approximately meet the desired specification. Finally, we will illustrate the design performance of the proposed method with two experiments. Simulation results reveal that the proposed scheme has a good design performance on the two-dimensional FIR digital filter.