Greenhouse cultivation stands out for the ongoing food production challenge due to its ability to maintain a specific microclimate and allow crops to grow under highly variable weather conditions. Since the energy resources in greenhouses are limited, an energy management system based on evolving fuzzy prediction intervals is proposed to correctly define a proper irrigation and water extraction schedule, assuming a limited amount of energy stored. The energy management system schedules the crop irrigation to fulfill a defined daily irrigation volume while managing the water extraction according to the future photovoltaic power. In addition, evolving fuzzy prediction intervals are used to forecast the photovoltaic power and estimate its worst-case scenario from the interval's lower bound. With this information, the energy management system can be implemented using a robust model predictive controller. The proposed controller is tested assuming a reduction in the real solar power the system receives, which resembles a change in the system dynamics due to shadows and dust. Then, the controller's performance is compared against conventional fuzzy prediction intervals. Simulation results show that the proposed energy management system fulfills the reference irrigation and achieves a 33% higher state of energy and 14% higher water availability, on average, during the simulation. Thus, the proposal is better prepared for energy and water shortages due to the controller's robust approach and the model's evolving nature.

Due to global warming and population growth, preserving and guaranteeing clean water and electricity access has become harder. For this purpose, Energy-water microgrids (EWMGs) have been proposed to manage both resources efficiently. In these systems, resource management is traditionally performed at the tertiary control level, on large time windows, whereas integration of renewable energy sources requires faster controllers. Several works proposed moving energy cost management to the secondary control level as a solution, achieving quick responses to perturbations. Inspired by this idea, We propose to solve the water-energy co-optimization at a secondary control level timescale, using the Karush-Khun-Tacker (KKT) conditions of the centralized economic dispatch (ED) of an EWMG. The proposal is validated through simulation, achieving an 11% operational cost reduction. While our simulations were executed on only one type of EWMG topology, the approach presented can be generalized to any topology.

This paper proposes a distributed predictive secondary controller that restores frequency deviations and handles the active power sharing of multiple generation units in a sea harbor microgrid. The proposed control strategy is based on a multi-objective optimization framework, where each distributed generation unit is handled by its own predictive optimization problem. The main contribution of this approach is the avoidance of weighting factors on each optimization problem, which in most works in the literature have to be tuned for the specific application case. Instead, in this proposal the control actions applied to the microgrid are determined by the obtention of a Pareto front defined in terms of the different control objectives in a microgrid. This controller is validated via simulation by connecting and disconnecting loads in a sea harbor microgrid model. The reported results confirm that the proposed distributed controller based on multi-objective optimization can handle the operation of multiple generation units while complying with the frequency restoration and power consensus condition.

Microgrids (MGs) are sustainable solutions for rural zone electrification that use local renewable resources. However, only careful planning at the start of an MG project can ensure its future optimal operation. In this paper, a novel methodology for MG planning by using the uncertainty characterization of renewable resources and demand is presented. Additionally, a model of electricity consumption is proposed and applied in an isolated rural community. In such communities, consumption patterns typically need to be derived as model inputs because consumption measurements are not available for the planning stage. To obtain these inputs, clustering algorithms based on self-organizing maps (SOMs) and fuzzy c-means are used to classify the families of the community given sociodemographic information obtained via surveys. Subsequently, Markov chains (MCs) are employed to generate consumption patterns based on consumption measurements in some dwellings and surveys applied to the community. The nonlinearities and uncertainties associated with renewable resources and consumption are modeled by using prediction interval (PI) models. These PI models provide the required consumption and generation scenarios for deriving the optimal sizing and topological information to address the MG planning problem. The results of the robust planning approach based on scenarios are useful at the feasibility and design phases of an MG project. The proposed methodology is successfully applied to MG planning for a rural Mapuche community, where a conservative criterion was considered to minimize the investment risk. This criterion corresponds to the worst-case scenario in which the demand increases by 19.9% compared to that of the baseline scenario and a lower energy cost is obtained. However, the net present cost and operational costs increase by 14% and 11.75% compared to those of the baseline scenario, respectively.

Water resources have a vital role in maintaining crops' survival in agricultural activities. Still, this resource's availability is limited and strongly affected by the climatic conditions of the area where the crops are located. Because of this situation, Energy-Water Management Systems have been implemented to optimize the use of resources for operating an irrigation system, while avoiding the over-extraction of water from the aquifer. However, these controllers require accurate future predictions of the climatic variables, which usually have a strong stochastic behavior. Therefore, this work proposes a Robust Energy-Water Management System based on prediction interval models to handle the uncertainty generated by the behavior of these climatic signals and the scarcity of accurate data for training the models. This work first analyzes available data from different sources to select the proper dataset for each climatic variable. Then, prediction intervals based on deep learning and fuzzy models are constructed for modeling solar radiation, air temperature, and precipitations. Using the information provided by the intervals, the proposed robust predictive controller is implemented and compared for two different cases, a conventional controller that uses the expected values of the climatic variables and a hypothetical ideal case where the controller knows the future with exact precision. Simulation results show that the prediction intervals based on deep learning and fuzzy models can reach good results for modeling these meteorological signals. Additionally, the proposed controller succeeds in maintaining the crops alive with the water available while getting a crop's profit close to the value achieved in the hypothetical ideal case. One of the main conclusions of this work is the importance that has the process of adequately analyzing different interval methods and data sources for achieving accurate models for climatic variables. Also, the quality of the interval models has an essential role in the controller for reaching results close to the ideal case, despite the stochasticity and uncertainty of the predicted signals.

This paper proposes a two-level hierarchical energy management system (EMS) with demand side management (DSM) capabilities for grid-connected microgrids (MGs). The proposed strategy is based on model predictive control (MPC) with prediction intervals obtained through the fuzzy numbers method. While the Main Grid level EMS aims for auto-consumption within the MG, i.e., minimise the energy drawn for the main grid, the Microgrid level tracks power and consumption references, sent from the higher level, to manage the MG resources and the load consumption. Furthermore, fuzzy prediction intervals are used to determine the best-case and worst-case scenarios of operation and modify the load profile while the overall load during the MG operation is maintained. Operation data for generation and consumption from a real urban community is used to validate the performance of the proposed EMS. The results show that the proposed hierarchical EMS with DSM and an adequate prediction case can reduce weekly costs while maintaining overall consumption and a healthy battery usage compared to an EMS that has no way to modify the load. This concludes that a microgrid can improve its performance with the correct predictions and the commitment of the consumers.

A robust model predictive control strategy based on fuzzy and neural prediction intervals is proposed to implement a greenhouse’s water and energy management system. The implementation of this model predictive control aims to optimize the energy use when controlling the irrigation process of crops based on the resources available in the greenhouse. In the dynamics considered for the greenhouse, the amount of energy available for the system’s operation is directly affected by climate conditions, such as ambient temperature and solar irradiance. Thus, the uncertainty associated with the stochastic behavior of these external disturbances can produce problems when deciding the optimal planning of energy use. Due to that, this work proposes to characterize these external signals by using prediction intervals based on fuzzy models and neural networks. Then, according to the information provided by the prediction intervals, the controller can now consider the worst-case scenarios for the energy available in the optimization problem solved by the predictive control strategy. Simulation results compare the performance of different prediction interval methods, showing their effectiveness for approximating the future behavior of the solar irradiance and ambient temperature and characterizing their uncertainty. Then, the proposed robust controllers based on the best intervals are compared with a deterministic model predictive control to show the proposal’s improvements in battery energy management.

This paper proposes a distributed predictive secondary controller to tackle together frequency and voltage regulation, realize the economic dispatch and reactive power sharing of generation units in isolated AC microgrids. Contrary to most approaches, the proposed predictive controller achieves consensus objectives (economic dispatch of generation and reactive power sharing) with soft constraints (keep both frequency and average voltage within predefined bands instead of restoring them to their nominal values). Extensive simulation work validates the effectiveness of the predictive controller for communication problems and in the presence of plug-and-play scenarios.

A model predictive control strategy (MPC) based on fuzzy optimization is proposed in this work for a multi-room heating, ventilation and air conditioning (HVAC) system. The proposed strategy aims to minimize energy consumption, while requiring different thermal conditions for each room. For this system, the combination of MPC and fuzzy optimization arises as a suitable control strategy, due to the benefits given by the use of fuzzy constraints for the management of thermal comfort.The soft constraint scheme provided by the fuzzy optimization allows to reduce the power consumption of HVAC systems. This is achieved by allowing some constraint violations in specific cases where the proper operation of the system is not compromised. In this context, the main contribution of this work is the introduction of a new framework where the thermal requirements of several rooms can be managed by fuzzy constraints, which are handled as additional terms in the objective function of the MPC. The optimization problem of the MPC is nonlinear, and it is solved with a suitable particle swarm optimization (PSO) method. Simulations results show the effectiveness of the proposed controller to reduce the energy consumption compared with a classical MPC implementation, while maintaining constraint satisfaction in appropriate levels.

This paper presents a new methodology for Prediction Interval (PI) construction based on a modified Takagi-Sugeno fuzzy system trained with a joint Supervision loss function. Given a desired coverage level, this model is capable of providing predictions of the expected value of the system along with the interval bounds. This methodology is tested by simulation experiments using a dataset containing real temperature data from a rural community in southern Chile. The proposed model was compared with a state-of-the-art Takagi-Sugeno Fuzzy Numbers model. It was shown that the Joint Supervision method manages to obtain slightly superior results to the Fuzzy Numbers approach while greatly reducing the complexity of the training loss function. Additionally, since the proposed model was trained using Particle Swarm Optimization, further performance improvements could be made by employing gradient-based optimization algorithms, since they are compatible with the Joint Supervision loss function.

This paper presents an approach for prediction interval generation by training a LSTM neural network with a joint supervision Loss Function. The prediction interval model provides the expected value and the upper and lower bounds of the interval given a desired coverage probability. The prediction interval models based on LSTM networks are compared with the classical recurrent neural network approach and are tested using two case studies. The first case corresponds to the forecasting up to one day ahead of the demand profile of 20 dwellings from a town in the UK, and the second case corresponds to the net power from an energy community made up 30 dwellings with a 50% level of photovoltaic power penetration. By using LSTM networks as the backbone of the proposed architecture, high-quality intervals are obtained with a narrower interval width compared with the classical recurrent neural network approach. Furthermore, the information provided by the prediction interval based on the LSTM network could be used to develop robust energy management systems that, for example, consider the worst-case scenario.

Many people across Latin America still do not have access to reliable electricity. Although Chile exhibits a comparatively high electricity coverage, many barriers are still present for the development of sustainable energy supply solutions exploiting local renewable energy sources. To face this challenge, a coconstruction methodology is proposed, which considers a flexible and participatory design with continuous communication between the technical team of the project and the community, thus ensuring informed decision making around the project design. In this context, microgrid-based solutions offer an ideal opportunity to exploit the integration of energy sources adapted to the specific local characteristics. The coconstruction methodology allows the identification of local requirements, less often considered for design procedures based on a traditional approach, in a joint work with the communities so that the technological solution is tailored for it. Consequently, different technical solutions (design adaptations and innovations) have been proposed and developed under this framework, such as: energy management systems, demand response strategies, microgrid applications for Mapuche communities, microformers, a monitoring system that includes social aspects, and vehicle to grid for microgrids. This paper summarizes the experience of several microgrid projects in Chile, identifies risks, impacts, control actions, and discusses their replicability to the Latin American and the Caribbean region.

A Robust MPC strategy for Heating, Ventilation and Air Conditioning Systems (HVAC) is proposed in this work. The typical control objective of minimizing energy consumption while maintaining user comfort is considered in this work. Robust MPC is naturally suited for HVAC systems with the aforementioned control goal because it is a control strategy that considers process constraints and the optimization of a performance index, and explicitly handles uncertainty. In this system, the uncertainty comes from the ambient temperature and internal loads predictions, which are the main factors driving the thermal dynamics. Thus, effectively predicting their future behaviour and uncertainty aids for the quality of the control system. In this context, the main contribution of this work is the introduction of a new framework that uses fuzzy interval models for predicting bounds of uncertain variables in a Robust MPC formulation. These bounds are constructed so that the future values of the relevant variables are within them with a predefined probability. In this work, fuzzy interval models are trained and used to predict the future system disturbances, and these are in turn used to provide bounds for the predictions of room temperatures of the HVAC system. Simulation results show the effectiveness of the proposed strategy, in terms of yielding higher percentage of constraints satisfaction when compared to a classical MPC method. Additionally, it is shown that an appropriate compromise between the system performance and the rate of constraint satisfaction can be achieved by varying the coverage probability of the fuzzy interval models.

In this paper, a new prediction interval model based on a joint supervision loss function for capturing the uncertainties associated with the modeled phenomenon is described. This model provides the upper and lower bounds of the predicted values in accordance with the desired coverage probability, as well as their expected values. A benchmark problem is used to evaluate the proposed method, and a comparison with the neural network covariance method is performed. Additionally, the proposed method was applied to forecast the residential demand from a town in UK, considering the prediction interval performance for one-day ahead. The results show that the method is able to generate an interval with narrower width than the covariance method, and maintains the coverage probability. The information provided by the prediction interval could be used in the design of microgrid energy management systems.

This paper presents robust evolving cloud-based controller (RECCo) for the distributed solar collector field (DSCF). The main issue of the DSCF is that the primary energy source (variable) cannot be manipulated. Beside this, unpredictable changing of environmental conditions (outlet temperature, cloudiness, solar radiation) on the daily basis strongly influence the dynamics of the whole process. According to this, the RECCo controller is robust enough to cope with the high levels of uncertainty present in the DSCF plant. RECCo is a fuzzy rule-based type of controller based on parameter-free premise (IF) part while the PID-type control consequent is used. Algorithm starts with zero fuzzy rules (zero clouds in data space). During operation it evolves its structure (adding new data clouds) and adapts the PID parameters for each data cloud while preforming the control of the plant. This means that no a-priori knowledge of the controlled process is required. Moreover, the ability of the learning is tested on the different operating points which cover the majority of the operating range of the DSCF plant.

An important issue in the operation of isolated microgrids is how to properly address the uncertainties associated with renewable resources and loads. In this work, a new interval prediction model based on type-2 fuzzy systems is derived for capturing these uncertainties. This model provides the upper and lower boundaries of the predicted values with a certain coverage probability, as well as their expected values. This information could be used in the design of microgrid controllers, such as energy management systems. The proposed method is applied to forecast the wind power generation and the loads of the Huatacondo microgrid, which is located in northern Chile. A comparison with both linear and Takagi-Sugeno fuzzy interval models is also performed.

Microgrids are sustainable solutions for electrification of rural zones that can make use of their local renewable resources. In this paper, we propose a new method for microgrid planning which includes the effect of the uncertainties of the renewable resources explicitly. Fuzzy interval models are used because they can capture nonlinearities and systematically represent the uncertainties associated with renewable resources at a certain confidence level. Relying on interval fuzzy models and by considering a set of possible scenarios for the renewable resources, the solution to the microgrid planning problem is given through the optimal sizing and topology of the microgrid. This information, particularly the optimal sizes of generators and the economic analysis, is useful for the design phase of a microgrid project. The proposed methodology is applied to the microgrid planning of the rural Mapuche community, José Painecura, in Chile.

In this paper a novel predictive control scheme based on Takagi-Sugeno model whose consequences include second order terms is proposed. Fuzzy models are used in order to approximate the non-linear behavior present on industrial dynamic systems. Quadratic approximations are used in the consequences because several systems has restricted controllable regions in the states domain. Thus, even fuzzy models may not be enough for representing the system dynamics in that regions, producing unexpected closed loop-behavior and loss of performance. The main difference between the proposed scheme and the ones reported in the literature is that iterative procedures and/or point to point approximation is not required. Reducing the computational burden of the controller. A continuous stirred tank reactor is used for testing the proposed control scheme.

This paper presents a model comparison of a fixed speed wind turbine (FSWT) operating on a real wind farm. By relying on real data obtained from a wind farm operating in the Chilean Interconnected System, three different models are identified and analyzed. First, a phenomenological model based on physical principles governing the production of electricity from wind power is considered. This model is fine-tuned in accordance with practical considerations, such as wind correction factors. Then, a linear model and a Takagi & Sugeno (T&S) fuzzy model are identified. From the experimental results, the linear model is the simplest one, but also the one that presents the worst performance indexes. The best prediction capability is obtained with the T&S model; however, in terms of interpretability, the phenomenological model outperforms the other two black-box models.

A battery storage system connected to a micro-grid through a Voltage Source Inverter (VSI) can have at least two operating modes (dual-mode). In the first one the VSI is connected to a synchronous generator which could be considered as a large capacity grid or infinite busbar. The second operating mode is obtained when the VSI is feeding an isolated load and several small micro-sources. Therefore, for a proper operation of this system, the VSI has to be controlled in order to operate in the grid-connected or stand-alone modes, with seamless transfer between them. This paper proposes a master-slave control strategy for the main energy storage VSI operating in a microgrid. The proposed control strategy achieves seamless transfer between these two situations with minimal communication requirements between the VSI and the micro-grid controller, and allows the VSI to operate in parallel with virtually any type of generator or inverter. The control scheme is formulated for a micro-grid operated using conventional droop strategy. A control strategy is proposed for each of the VSI control levels. The control strategies have been experimentally tested in a full practical micro-grid implemented in a small village located in the Chilean Andes. Experimental results are presented and analyzed in this paper.

In this paper, two methods for generating the daily load profile and forecasting in isolated small communities are proposed. In these communities, the energy supply is difficult to predict because it is not always available, is limited according to some schedules and is highly dependent on the consumption behavior of each community member. The first method is proposed to be used before the implementation of the microgrid in the design state, and it includes a household classifier based on a Self Organizing Map (SOM) that provides load patterns by the use of the socio-economic characteristics of the community obtained in a survey. The second method is used after the implementation of the microgrid, in the operation state, and consists of a neural network with on-line learning for the load forecasting. The neural network model is trained with real-data of load and it is designed to stay adapted according to the availability of measured data. Both proposals are tested in a real-life microgrid located in Huatacondo, in northern Chile (project ESUSCON). The results show that the estimated daily load profile of the community can be very well approximated with the SOM classifier. On the other hand, the neural network can forecast the load of the community reasonably well two-days ahead. Both proposals are currently being used in a key module of the energy management system (EMS) in the real microgrid to optimize the real uninterrupted load for 24-hour energy supply service.

In the daily operation of a bus system, the movement of vehicles is affected by uncertain conditions as the day progresses, such as traffic congestion, unexpected delays, randomness in passenger demand, irregular vehicle dispatching times, and incidents. In a real-time setting, researchers have devoted significant effort to developing flexible control strategies, depending on the specific features of public transport systems. In this paper, we propose a control scheme for the operation of a bus system running along a linear corridor, based on expert rules and fuzzy logic. The parameters of the fuzzy controllers were tuned through a particle swarm optimization (PSO) algorithm. That is, the control strategies aim at keeping regular headways between consecutive buses, with the objective of reducing the total waiting time of passengers. The proposed control systems rely on measures of the position of each bus, which are easy to obtain and implement by means of emerging automatic vehicle location devices through Global Positioning System (GPS) technology. The utilized strategies are holding, stop-skipping, and the integration of both. After tuning the controller parameters, we conducted several simulation tests, obtaining promising results in terms of savings in waiting times with the implementation of the proposed rules, noting that the best performance occurred when fuzzy rules are included. The methodology has great impact, and it is easy to implement due to its simplicity.

In this paper, a dynamic vehicle routing problem (DVRP) is solved based on hybrid predictive control strategy with an objective function that includes two dimensions: user and operator costs. To handle some undesired assignments for the users, a new objective function is designed, able to carry out the fact that some users can become particularly annoyed if their service is postponed. Genetic algorithms are proposed for efficiently solving the DVRP. Fuzzy clustering is applied for computing trip patterns from historical data under more realistic scenarios. An illustrative experiment through simulation of the process is presented to show the potential benefits (mainly for users) of the new design.

In this work, we propose a combined approach based on Particle Swarm Optimization (PSO) and Sequential Quadratic Programming (SQP) for solving the real-time optimization problem appearing in the hybrid predictive control of systems with continuous and discrete inputs. The conventional PSO for continuous problems is mixed with binary PSO for handling discrete variables, and after iterations of the algorithm have ended, SQP is applied for refining the continuous variables. This new method (PSO+SQP) is favorably compared with a conventional non-linear optimization techniques (based on Branch and Bound and Explicit Enumeration) and PSO. Also, a multi-objective approach is considered as a tuning method for the population size and number of iterations for PSO. All these algorithms are applied for the temperature control of a Batch Reactor.

In this work, a new application for robot hand identification using type-2 fuzzy intervals is presented. Type-2 identification method by using only input-output data determines the parameters of upper and lower membership functions of interval sets. The identification method was successfully applied to the modeling of a robot hand Moreover, we show how a type-2 fuzzy model improves the description of the robot hand in comparison with type-1 fuzzy model, if uncertain inputs are considered.

In the paper, the hybrid predictive control based on a fuzzy model is presented. The identification methodology for a nonlinear system with discrete state-space variables by combining fuzzy clustering and principal component analysis is proposed. The fuzzy model is used for hybrid predictive control design where the optimization problem is solved by the use of genetic algorithms. An illustrative experiment on a hybrid tank system is conducted to present the benefits of the proposed approach.

Particle Swarm Optimization (PSO) is proposed as an efficient tool for the design of fuzzy predictive control (FPC) strategies. The performance of the proposed method is evaluated in terms of accuracy and computational time. PSO is compared with other evolutionary algorithms such as simple genetic algorithms and niching genetic algorithms. Simulation results that validate the proposed FPC-PSO scheme are presented for a benchmark non-linear series.

In this paper, a new selection method of relevant input variables of Takagi & Sugeno models is proposed. Also, a general identification methodology is described. The proposed method is exemplified using a benchmark problem and also a data set from a combined cycle power plant boiler simulator

This work presents the application of fuzzy predictive control to a solar power plant. The proposed predictive controller uses fuzzy characterization of goals and constraints, based on the fuzzy optimization framework for multi-objective satisfaction problems. This approach enhances model based predictive control (MBPC) allowing the specification of more complex requirements. A brief description of the solar power plant and its simulator is given. Basic concepts of predictive control and fuzzy predictive control are introduced. Two fuzzy predictive controllers using different membership functions are designed for a solar power plant, and they are compared with a classical predictive controller. The simulation results show that the fuzzy MBPC formulation, based on a well proven successful algorithm, gives a greater flexibility to characterize the goals and constraints than classical control.

In this paper, a novel control strategy based on neural networks is proposed in order to reduce the computation effort of a nonlinear predictive controller. The proposed method is favorably compared with the nonlinear predictive controller and approximated predictive controller based on neural networks. Also, the control strategies are designed and evaluated by simulation tests and in real-time for a laboratory tank system.

In this paper, a supervisory optimal control problem is presented. Fuzzy modelling is used to represent the non-linearity of the process and two alternative fuzzy predictors are described in order to solve the optimisation problem at the supervisory level. The proposed fuzzy predictive controllers are successfully applied to a benchmark problem.