Enric Sala

Constrained-size torque maximization in SynRM machines by means of genetic algorithms

Synchronous Reluctance Motors have always been an alternative to more mainstream machines such as the Permanent Magnet Synchronous Motor, but until recently they have not found their right place in industrial applications. This progressing adoption begins with the replacement of the current solutions, but this presents the design challenge of finding a surrogate which conforms to the specifications for a given application. In order to overcome this challenge, an evolutionary design methodology for SynRM is presented. The proposed approach uses a set of design constraints to maximize the mechanical power of the motor taking into account the specified rated speed. Since the calculation of the torque of the motor is critical, an iterative method for the evaluation of iron losses has been introduced. Finally, the proposed approach is validated by means of FEM simulation and the calculation of the efficiency map of the results.

Smart multi-model approach based on adaptive Neuro-Fuzzy Inference Systems and Genetic Algorithms

A model of power demand represents the foundation of any intelligent Energy Management System, and its accuracy is the key factor determining the performance of such system. In order to improve the accuracy of the modeling process, a multi-model approach based on a Hierarchical Clustering of similar load behaviors is presented. The clustering algorithm joins similar data subsets in groups that are modelled separately using Adaptive Neuro-Fuzzy Inference Systems. Thus, each of the obtained models addresses only the characterization of one behavior, which provides better accuracy than classical approaches based on a single model, in addition to being easier and faster to train. During the training process of the models, an input selection technique based on Genetic Algorithms is proposed to search and select the best combination of inputs. The use of search algorithms allows to reduce the complexity of this task while maintaining the system performance, which represents a significant time saving of expert staff. The proposed approach is validated by means of experimental data from an automotive manufacturing plant. In addition to improving the forecasting accuracy, this methodology automates the segmentation of the load profiles into models and the selection of their inputs, as well as improving parallelization to effectively reduce the computation time.

Multiobjective Optimization of Multi-Carrier Energy System Using a Combination of ANFIS and Genetic Algorithms

This paper presents a novel method for the energy optimization of multi-carrier energy systems. The presented method combines an adaptive neuro-fuzzy inference system, to model and forecast the power demand of a plant, and a genetic algorithm to optimize its energy flow taking into account the dynamics of the system and the equipment’s thermal inertias. The objective of the optimization algorithm is to satisfy the total power demand of the plant and to minimize a set of optimization criteria, formulated as energy usage, monetary cost, and environmental cost. The presented method has been validated under real conditions in the car manufacturing plant of SEAT in Spain in the framework of an FP7 European research project.

Disaggregation of HVAC load profiles for the monitoring of individual equipment

This paper presents a load disaggregation method for the monitoring and supervision of the load profiles of individual equipment in an HVAC installation. The method takes advantage of the wealth of sensor and actuation information found in Building Energy Management Systems in order to find correlations between the state of operation of each machine and the power demand of the installation. This enables to model the individual power of the equipment on account of their state, and in combination with other support variables that influence their load demand, such as weather conditions. The resulting array of equipment models can be evaluated in real-time to infer the expected power consumption of each machine. Then, allowing the tracking of their individual power consumption while at the same time significantly lowering the cost of the acquisition and monitoring infrastructure, because a single power meter can be used to accurately monitor several machines when following this approach. The presented method has been validated by means of experimental data from a pilot plant where the complete system has been implemented.

Enhanced load forecasting methodology by means of probabilistic prediction intervals estimation

The improvement of the forecasting accuracy for prediction of future loads has been object of exhaustive study in the recent years, to the point that a wide variety of methodologies which have been proved to be valid and practical exists. However, most methodologies for demand forecasting do not handle uncertainties of the resulting model, which leads to a nonproper interpretation of the forecasted outcomes. In this context, this work presents a novel load forecasting methodology in order to quantify the model uncertainties and complement the resulting information by means of adaptive confidence intervals. First, an input selection technique based on Genetic Algorithms is used to select the best combination of inputs in order to obtain a state-of-the-art model by means of Adaptive Neuro-Fuzzy Inference Systems. Then the data space is analyzed in terms of error probability of the model outcomes. The principal component analysis is used to visualize the error probability in a 2-D map. Finally, an Artificial Neural Network is used to perform the identification of the error probability associated to new measurements. In conjunction with the forecasting model, the proposed classifier extends the resulting information with an adaptive confidence intervals and its probability distribution. The effectiveness of this enhanced load forecasting methodology has been verified by experimental data obtained from an automotive plant.

Time series forecasting by means of SOM aided Fuzzy Inference Systems

The forecast of industrial process time series represents a critical factor in order to assure a proper operation of the whole manufacturing chain, as it allows to act against any process deviation before it affects the final manufactured product. In this paper, in order to take advantage from process relations and improve forecasting performance, a prediction method based in Adaptive Neuro Fuzzy Inference System (ANFIS) and Self-Organizing Maps is presented. The novelties of the proposed method are based on considering, as an input of an ANFIS model, the interrelations of process variables regarding the signal that wants to be forecasted, by means of topology preservation SOM. An experimental study performed with real industrial data from a cooper manufacturing plant indicated the suitability of the proposed method in time series forecasting applications.

Short-term load forecasting using Cartesian Genetic Programming: An efficient evolutive strategy: Case: Australian electricity market

Currently, the Cartesian Genetic Programming approaches applied to regression problems tackle the evolutive strategy from a static point of view. They are confident on the evolving capacity of the genetic algorithm, with less attention being paid over alternative methods to enhance the generalization error of the trained models or the convergence time of the algorithm. On this article, we propose a novel efficient strategy to train models using Cartesian Genetic Programming at a faster rate than its basic implementation. This proposal achieves greater generalization and enhances the error convergence. Finally, the complete methodology is tested using the Australian electricity market as a case study.

Optimal control of energy hub systems by use of SQP algorithm and energy prediction

This paper presents an energy optimization methodology applied on industrial plants with multiple energy carriers. The methodology combines an adaptive neuro-fuzzy inference system to calculate the short-term load forecasting of a plant, and the sequential quadratic programming algorithm to optimize its energy flow. Furthermore, the mathematical models of the plants equipment are considered into the optimization process, in order to calculate the dynamic system response and the equipments inertias. The final algorithm optimizes the operation of the plant in order to satisfy the energy demand, minimizing several optimization criteria. The methodology has been tested and evaluated in an automotive factory plant using real production and consumption data.

Industrial process monitoring by means of recurrent neural networks and Self Organizing Maps

Industrial manufacturing plants often suffer from reliability problems during their day-to-day operations which have the potential for causing a great impact on the effectiveness and performance of the overall process and the sub-processes involved. Time-series forecasting of critical industrial signals presents itself as a way to reduce this impact by extracting knowledge regarding the internal dynamics of the process and advice any process deviations before it affects the productive process. In this paper, a novel industrial condition monitoring approach based on the combination of Self Organizing Maps for operating point codification and Recurrent Neural Networks for critical signal modeling is proposed. The combination of both methods presents a strong synergy, the information of the operating condition given by the interpretation of the maps helps the model to improve generalization, one of the drawbacks of recurrent networks, while assuring high accuracy and precision rates. Finally, the complete methodology, in terms of performance and effectiveness is validated experimentally with real data from a copper rod industrial plant.

Intelligent monitoring of HVAC equipment by means of aggregated power analysis

The increasing ubiquity of sensing and metering devices in buildings is a gateway of opportunities for the analysis of their behavior and the detection of anomalies or sub-optimal performance. In particular, the instrumentation of HVAC equipment, necessary for its monitoring and control, may be used in order to supervise its operation at a finer level. The intelligent supervision methodology proposed in this paper allows the accurate overseeing of the power consumption of HVAC equipment by means of establishing the relationship between the power consumed and the operating status of the installation, and its individual machines. The accurate correlation between the instantaneous power and the available control or state signals of the equipment, simultaneously with other support variables, allows detecting malfunctions or deviations from their nominal operation. First, a model of the power demand of the installation is obtained by means of a training function. Afterwards, the model can be applied in real-time over new samples in order to check if the power demand corresponds to the state of operation of the installation. In addition to the accurate tracking of the power demand, the chosen approach allows to monitor the installation with a single power meter, therefore decreasing the cost of implementation. Finally, this methodology has been validated by means of experimental data from a pilot plant where the complete system has been implemented.