Fabrice Locment

Building Integrated Photovoltaic System With Energy Storage and Smart Grid Communication

The utility grid challenge is to meet the current growing energy demand. One solution to this problem is to expand the role of microgrids that interact with the utility grid and operate independently in case of a limited availability during peak time or outage. This paper proposes, for urban areas, a building integrated photovoltaic (BIPV) primarily for self-feeding of buildings equipped with PV array and storage. With an aim of elimination of multiple energy conversions, a DC network distribution is considered. The BIPV can supply a tertiary building at the same time as PV array may produce power through a hierarchical supervision able to exchange messages with the smart grid and metadata. The hierarchical control is designed as an interface to expand the system ability for advanced energy management control having regard to the grid availability and users commands. It consists of four layers: human-machine interface, prediction, cost management, and operation. The operation layer, implemented in an experimental platform, takes into account the grid supply power limits and constrains the DC load. The experimental results validate the approach that may be a solution for the future smart grid communication between BIPV and utility grid.

Intelligent DC Microgrid With Smart Grid Communications: Control Strategy Consideration and Design

Aiming at photovoltaic (PV)-storage urban building integrated system, this paper proposes a DC microgrid with multi-layer control and smart grid communications. The paper focuses on power balancing, with load shedding and PV constrained production, and takes into account the grid availability and grid vulnerability by smart grid messages. The system behavior modeling by MATLAB Stateflow leads to the whole control strategy design, which concerns the power balancing and imposed power limits by the utility grid, while providing interface for energy management. Experimental results evaluate the feasibility of the proposed control strategy. As further development of this control design, an intelligent multi-layer supervision is suggested. This supervision, able to exchange data with the smart grid, deals with the end-user demand, forecast of photovoltaic production, prediction of load consumption, and energy management. The major technical contribution of this paper is linked to the proposed control design that permits better DC microgrid integration (avoids undesired injection, mitigates fluctuations in grid power, reduces grid peak consumption) and provides possibility to reduce the negative impact on the utility grid thanks to the supervision interface. The power balancing control interface provides possibility for advanced energy management with low speed communication.

DC Load and Batteries Control Limitations for Photovoltaic Systems. Experimental Validation

This study first presents an experimental control strategy of photovoltaic (PV) system composed of: PV array, dc-dc power converters, electrolytic storage, and programmable dc electronic load. This control aims to extract maximum power from PV array and manages the power transfer through the dc load, respecting the available storage level. The designed system allows simultaneously the supply of a dc load and the charge or the discharge of the storage during the PV power production. The experimental results obtained with a dSPACE 1103 controller board show that the PV stand-alone system responds within certain limits that appear as soon as one of the storage thresholds is reached: either loss of energy produced, or insufficient energy toward the load. In urban area, it is proposed to overcome these limitations by connecting the utility grid with the PV system while maintaining the priority for self-feeding. The experimental results of this PV semi-isolated system are shown and discussed. For this first approach, the goal was to verify the technical feasibility of the suggested system controls. The final results are energetically relevant.

A simple PV constrained production control strategy

Photovoltaic (PV) panels usually work with a Maximum Power Point Tracking (MPPT) algorithm to produce as much power as possible. Their output power changes mainly according to the solar irradiation, so they are unpredictable power sources. For grid-connected PV systems, the PV MPPT production could be strongly fluctuating due to possible significant variations of solar irradiation, which could be an issue for grid power quality with consideration of future PV penetration level. On the other hand, for stand-alone PV systems, which usually work with energy storage, the MPPT production would cause extra problems in power balancing when the storage reaches its upper state of charge limit. In these cases, the PV production should be constrained. This paper proposes a simple PV constrained production control strategy, which leads to control the PV power at any level within the MPPT production ability. If the solar irradiation decreases and PV MPPT power is not able to output the desired constrained power, the control strategy continues to operate PV system with MPPT algorithm. The transition between MPPT and constrained mode is continuous and seamless; the two modes operate separately. Experimental results validate the proposed control strategy.

Power management and optimization for isolated DC microgrid

This paper presents a photovoltaic-storage-microturbine based DC microgrid, with multi-layer supervision which optimizes efficiently operation while satisfying constraints. Based on forecast data and using a mixed integer linear programming optimization, a predictive power flow is obtained which leads to an optimization-based strategy for real-time power balancing. Experimental test proves the feasibility of the supervisory control and demonstrates the problem formulation of minimizing total energy cost.

Electric vehicle charging system with PV Grid-connected configuration

This paper presents an experimental control strategy of electric vehicle charging system composed of photovoltaic (PV) array, converters, power grid emulator and programmable DC electronic load that represents Li-ion battery emulator. The designed system can supply the battery at the same time as PV energy production. The applied control strategy aims to extract maximum power from PV array and manages the energy flow through the battery with respect to its state of charge and taking into account the constraints of the public grid. The experimental results, obtained with a dSPACE 1103 controller board, show that the system responds within certain limits and confirm the relevance of such system for electric vehicle charging.

Multi-source power generation system in semi-isolated and safety grid configuration for buildings

Due to technical constraints of major incidents related to connecting distributed green sources with the public grid, we propose, in urban areas, a semi-isolated and safety system for self-feeding of buildings equipped with renewable electricity. With an aim of elimination of multiple energy conversions, a direct current (DC) network distribution is considered. The semi-isolated and safety DC network is studied using a multi-source system (photovoltaic panels, storage, and public grid) and numerical simulation results are given. The results focus on the redefinition of the electrical network for buildings equipped with renewable sources, with the proposal of a semi-isolated grid. These results confirm the relevance of such network in urban areas, and can map out perspectives on integrated and optimized energy management of the building.

A simple experimental prediction model of photovoltaic power for DC microgrid

The photovoltaic (PV) power prediction is necessary for dealing with the grid vulnerability by smart grid messages, and could reduce the negative impact due to the unpredictable power injection into the grid. For a DC microgrid with PV-storage integrated system, this paper aims to present a simple experimental prediction model of a PV panel. The proposed model is based on indoor measurements and allows predicting the maximum output power of the PV panel at given irradiance and cell temperature. Compared to two other models, a linear power model and a single-diode model, experimental tests prove the reliability of the proposed model and validate the method.

Day-ahead microgrid optimal self-scheduling: Comparison between three methods applied to isolated DC microgrid

This paper presents three different methods to optimize a day-ahead self-scheduling for an isolated DC microgrid operation. Based on forecast data, a multi-objective cost function is formulated aiming to minimize the total energy cost by reducing the micro-turbine fuel consumption, avoiding photovoltaic power limitation and load shedding, while respecting the storage parameters and microgrid operation constraints. The problem is solved with two different optimization algorithms and one rule-based algorithm. The comparison is made on the total energy cost and computational cost for each proposed approach. The results prove that the mixed integer linear programing optimization permits to obtain the lower total energy cost with a reasonable computational cost.

Original article: Power flow Petri Net modelling for building integrated multi-source power system with smart grid interaction

This paper presents an energy management modelling of a multi-source power system composed of photovoltaic (PV) array, storage and power grid connection, and taking into account messages from smart grid. The designed system can supply a tertiary building at the same time as PV may produce energy. The control strategy aims to manage the power flow through the load with respect to its power demand and public grid constraints. The proposed energy management modelling is based on interpreted Petri Nets formalism. The system is tested using a simulation Stateflow model and responds within certain limits. The results show that this approach is valid and can be a solution for the future smart grid communication between buildings and public grid and may contribute to a better balance between production and consumption and future energy management.