Stochastic energy management of a microgrid incorporating two-point estimation method, mobile storage, and fuzzy multi-objective enhanced grey wolf optimizer.

In this study, the stochastic energy management, and scheduling of a renewable microgrid involving energy sources and dynamic storage is performed considering energy resource and demand uncertainties and demand response (DR) using the two-point estimation method (2 m + 1 PEM). The three-dimensional objective function is defined as maximizing the renewable hosting capacity and minimizing the operation cost, and emission cost minimization. The decision variables include installation location and size of the renewable resources and mobile energy storage system (MESS), determined using a multi-objective enhanced grey wolf optimizer (MOEGWO) improved based on the logistic chaotic mapping integrated with fuzzy decision-making approach. The simulations are implemented for several cases of employing MESS, DR, and uncertainties to investigate the proposed approach's efficacy. The MOEGWO performance is confirmed to solve the ZDT and CEC'09 functions according to some well-known algorithms. Then, the performance of the MOEGWO is evaluated on the stochastic energy management and scheduling of the renewable microgrid. The results indicate that considering the dynamic MESS causes reducing the operation and emission costs by 23.34% and 34.78%, respectively, and increasing the renewable hosting capacity by 7.62% in contrast to using the static MESS. Also, the stochastic problem-solving considering uncertainties showed that operation and emission costs are raised, the renewable hosting capacity is decreased, and the uncertainty impact is reduced in the condition of DR application. So, the results validated the proposed methodology's effectiveness for minimizing the operation and emission costs and maximizing the renewable hosting capacity. Moreover, the superior capability of the MOEGWO is confirmed in comparison with the multi-objective particle swarm optimization to obtain lower operation and emission costs and higher renewable hosting capacity.

Fault detection and isolation in the challenging Tennessee Eastman process by using image processing techniques.

The early fault detection and isolation in industrial systems is a critical factor in preventing equipment damage. In the proposed method, instead of using the time signals of sensors, the 2D image obtained by placing these signals next to each other in a matrix has been used; and then a novel fault detection and isolation procedure has been carried out based on image processing techniques. Different features including texture, wavelet transform, mean and standard deviation of the image accompanied with MLP and RBF neural networks based classifiers have been used for this purpose. Obtained results indicate the notable efficacy and success of the proposed method in detecting and isolating faults of the Tennessee Eastman benchmark process and its superiority over previous techniques.

Process fault isolation based on transfer entropy algorithm.

Complexity of industrial plants and their stringent environmental and safety regulations have necessitated early detection and isolation of process faults. All the existing fault isolation methods can be categorized into two general groups: model-based and data-based. Transfer entropy is a data-based method for measuring propagation direction of disturbance and finding its root cause. In this paper, a new transfer entropy-based method is proposed to isolate different process faults. The novelty of this paper lies in using the transfer entropy idea to generate distinct patterns of information flow among process variables, recognize their correlations in the context of the transferred information in any abnormal condition, and finally isolate different process faults. The experimental results clearly demonstrate the superiority of the proposed method to the conventional methods.