Planning strategies for distributed photovoltaic based on fuzzy cognitive maps

The increased presence of photo-voltaic (PV) power generation in distribution grids has influences on the security of supply, in particular voltage stability. One of the commonly accepted control approach to remediate over-voltage situations is the absorbing of reactive power by the PV inverters. An alternative approach is to introduce energy buffering devices (e.g. batteries) into the distribution grid and to control real power charged respectively discharged from the batteries. We have simulated an examplary rural distribution grid with PV generation and energy buffering in order to study the effect on voltage distribution and power flows.

Distributed photovoltaic (PV) access to distribution network will affect the line loss and voltage of the system, and affect the reliability and economic operation of the distribution system. Therefore, in this study, firstly, node analysis models of distribution network system with four different typical load distribution is established. Secondly, the changes of voltage and network loss caused by distributed PV access to distribution network systems with four typical load distribution are analysed, and the formula of optimal access position and optimal access capacity of distributed PV system under different load distribution system is derived. Thirdly, on the basis of the work data of a distribution line with 43 distributed PV access in Tongchuan city, Shaanxi Province, China, the influence of distributed PV power generation on line loss and voltage of distribution network under different weather conditions are analysed. Finally, an optimal PV access scheme for distribution network based on ant colony algorithm is proposed to ensure the reliable and economical operation of distribution network after distributed PV access. The simulation model of distribution system with distributed PV access is established under Matlab, and the simulation results show the correctness and effectiveness of the proposed formula and the optimisation scheme.

Fuzzy Cognitive Maps are recognized knowledge modeling tool. FCMs are visualized with directed graphs. Nodes represent information, edges represent relations within information. The core element of each Fuzzy Cognitive Map is weights matrix, which contains evaluations of connections between map's nodes. Typically, weights matrix is constructed by experts. Fuzzy Cognitive Map can be also reconstructed in an unmanned mode. In this article authors present their own, new approach to time series modeling with Fuzzy Cognitive Maps. Developed methodology joins Fuzzy Cognitive Map reconstruction procedure with moving window approach to time series prediction. Authors train Fuzzy Cognitive Maps to model and forecast time series. The size of the map corresponds to the moving window size and it informs about the length of historical data, which produces time series model. Developed procedure is illustrated with a series of experiments on three real-life time series. Obtained results are compared with other approaches to time series modeling. The most important contribution of this paper is description of the methodology for time series modeling with Fuzzy Cognitive Maps and moving windows.

Long term supporting schemes for photovoltaic (PV) system installation have led to accommodating large numbers of PV systems within load pockets in distribution grids. High penetrations of PV systems can cause new technical challenges, such as voltage rise due to reverse power flow during light load and high PV generation conditions. Therefore, new strategies are required to address the associated challenges. Moreover, due to these changes in distribution grids, a different response behavior of the distribution grid on the transmission side can be expected. Hence, a new equivalent model of distribution grids with high penetration of PV systems is needed to be addressed for future power system studies. The thesis contributions lie in three parts. The first part of the thesis copes with the PV modelling. A non-proprietary PV model of a three-phase, single stage PV system is developed in PSCAD/EMTDC and PowerFactory. Three different reactive power regulation strategies are incorporated into the models and their behavior are investigated in both simulation platforms using a distribution system with PV systems. In the second part of the thesis, the voltage rise problem is remedied by use of reactive power. On the other hand, considering large numbers of PV systems in grids, unnecessary reactive power consumption by PV systems first increases total line losses, and second it may also jeopardize the stability of the network in the case of contingencies in conventional power plants, which supply reactive power. Thus, this thesis investigates and develops the novel schemes to reduce reactive power flows while still keeping voltage within designated limits via three different approaches: 1. decentralized voltage control to the pre-defined set-points 2. developing a coordinated active power dependent (APD) voltage regulation Q(P) using local signals 3. developing a multi-objective coordinated droop-based voltage (DBV) regulation Q(V) using local signals In the third part of the thesis, furthermore, a gray-box load modeling is used to develop a new static equivalent model of a complex distribution grid with large numbers of PV systems embedded with voltage support schemes. In the proposed model, variations of voltage at the connection point simulate variations of the model’s active and reactive power. This model can simply be integrated into load-flow programs and replace the complex distribution grid, while still keeping the overall accuracy high. The thesis results, in conclusion, demonstrate: i) using rms-based simulations in PowerFactory can provide us with quite similar results using the time domain instantaneous values in PSCAD platform; ii) decentralized voltage control to specific setpoints through the PV systems in the distribution grid is fundamentally impossible due to the high level voltage control interaction and directionality among the PV systems; iii) the proposed APD method can regulate the voltage under the steady-state voltage limit and consume less total reactive power in contrast to the standard characteristic Cosf (P) proposed by German Grid Codes; iv) the proposed optimized DBV method can directly address voltage and successfully regulate it to the upper steady-state voltage limit by causing minimum reactive power consumption as well as line losses; v) it is beneficial to address PV systems as a separate entity in the equivalencing of distribution grids with high density of PV systems.

Unregulated utilization of renewable generation including residential photovoltaic (PV) systems can have a significant impact on load characteristics in distribution networks. For improving PV generation capabilities, power quality aspects have to be coordinated with present load characteristics. This paper discusses the harmonic content of PV generation and the influence to power quality indicators in residential distribution networks. PV generation measurement results including current harmonic amplitude and phase angle values are presented. Results of different modelling scenarios are analysed and a simplified model of harmonics in PVs is offered. The results of the study showed a moderate additional harmonic distortion in residential load current and voltage distortion at the substation’s busbar when PVs were added. Novelty of the paper is that harmonic current values at higher orders are presented and analysed. The results pointed out in this paper could be further used for modelling the actual harmonic loads of the PVs in distribution networks.

The increasing penetration of solar photovoltaics (PVs) generation in distribution grids necessitates the need for optimal operation and scheduling of active/reactive resources for regulating the voltage along distribution feeders and reducing power consumption. In this paper, a mixed integer linear programming (MILP) risk-averse stochastic optimization model is proposed to co-optimize the traditional switching of capacitor banks and transformer tap along with PV and energy storage system (ESS) inverters. In day-ahead (DA) stage, traditional devices and purchased power of DA are scheduled. In real-time stage, the fast response inverters, ESS, and real-time market ensure sufficient active and reactive power support for distribution grids considering the conservation voltage reduction (CVR) plan. The CVR is integrated with the framework to reduce the energy consumption of voltage dependent loads by operating the grid close to the lower acceptable voltage ranges. The uncertainty of sudden changes in PV generation is represented by a Gaussian Mixture model (GMM). The generated uncertainty scenarios are reduced using an unsupervised fuzzy k-means method. Finally, the effectiveness of the proposed framework is verified using a modified version of the unbalanced IEEE 33-node system.

In this paper, the conventional Fuzzy Cognitive Maps (FCMs), which has already achieved success in many fields, are extended by using triangular fuzzy numbers (TFNs). The advantage of FCMs is that they are relatively easy to construct and parameterize and are capable of handling the full range of system feedback structure, including density-dependent effects. However, it is a well-known fact that there are some limitations inherent in FCM, such as lack of adequate capability to handle uncertain information and lack of enough ability to aggregate the information from different sources. Triangular fuzzy numbers which are represented by a triplet has the capacity to represent the uncertain relations between the concepts. In this context, the weight matrix representing the causal relations are enhanced to a fuzzy weight matrix that has TFNs as element. As a result of this improvement, the dynamic reasoning algorithm of the conventional FCM is improved for the use of the proposed novel FCM. The proposed FCM is presented via four simulations and the results are discussed. The results of the simulation study shows how easily the uncertain information can be represented and interpreted by the proposed FCM design methodology.

The demand for interpretable and accurate machine learning models continues to grow, especially in critical domains. The data-driven Fuzzy Cognitive Map (FCM) classifier is an interpretable and transparent decision-making method. Its core element, the weight matrix, is derived using predominantly population-based supervised learning methods which often suffer from degraded performance. Recent research has adopted gradient-based learning techniques to compete with the predictive performance of black-box models. Nonetheless, such methods modify foundational principles and compromise interpretability, highlighting the necessity to improve existing approaches. In this work, we introduce a novel learning and structural modeling method, termed Neural-FCM, which leverages deep neural networks and gradient descent to enhance the accuracy and robustness of FCM learning. Neural-FCM employs a hybrid network comprising both dense and convolutional layers and is trained using a categorical cross-entropy loss function specifically aligned with FCM reasoning. This hybrid model is trained to output instance-specific weight matrices for effective and targeted FCM inference, introducing structural adaptability, a feature not supported by previous static or globally optimized approaches. Focusing on generalization across domains, the Neural-FCM approach is evaluated on different classification tasks across six widely used public datasets and one proprietary medical dataset, consistently showing improved predictive performance. Notably, the comparative analysis against standard population-based FCM learning methods reveals consistent accuracy improvements, with gains of up to 34%. While less transparent gradient-based methods also yield improved accuracy, Neural-FCM demonstrates competitive or superior performance in most cases, with accuracy improvements ranging from 1 to 6% across different domains, while preserving the underlying interpretability. The performance enhancement and the use of instance-specific matrices contribute to the broader goal of developing gradient-based models that balance computational efficiency with the intrinsic FCM interpretability.

This study presents a survey on the most recent learning approaches and algorithms that are related to fuzzy cognitive maps (FCMs). FCMs are cognition fuzzy influence graphs, which are based on fuzzy logic and neural network aspects that inherit their main advantages. They gained momentum due to their dynamic characteristics and learning capabilities. These capabilities make them essential for modeling and decision-making tasks as they improve the performance of these tasks. An efficient number of learning algorithms for FCMs, by modifying the FCM weight matrix, have been developed in order to update the initial knowledge of human experts and/or include any knowledge from historical data in order to produce learned weights. The proposed learning techniques have mainly been concentrated on three directions: on the production of weight matrices on the basis of historical data, on adaptation of the cause-effect relationships of the FCM on the basis of experts' intervention, and on the production of weight matrices by combining experts' knowledge and data. The learning techniques could be categorized into three groups on the basis of the learning paradigm: Hebbian-based, population-based, and hybrid, which subsequently combine the main aspects of Hebbian-based- and population-based-type learning algorithms. These types of learning algorithms are the most efficient and widely used to train the FCMs, according to the existing literature. A survey on recent advances on learning methodologies and algorithms for FCMs that present their dynamic capabilities and application characteristics in diverse scientific fields is established here.

In last two decades there is great penetration of embedded generation in distribution grids. That penetration is present despite voltage level or connection point areas (rural, urban). Such high level of embedded generation is expected, taking into account need for reduction of greenhouse gas emissions and environmental protection. But, by increasing number of embedded generators, traditional distribution grids become active grids compared to traditionally passive grids. The concept of distribution grids with embedded generators is widely changed compared to passive grids in regards mostly to power flow, system protection and voltage quality. Voltage quality is very important factor that can be threatened after connection of PV plant in distribution grid. Very important fact about connection of PV plant on distribution grid can be connection point, point of common coupling (PCC). PV plant can have great influence on voltage quality in distribution grid considering PCC.

Impact of rapid PV fluctuations on power quality in the low-voltage grid and mitigation strategies using electric vehicles

High penetrations of photovoltaic (PV) systems within load pockets in distribution grids have changed pure consumers to prosumers. This can cause technical challenges in distribution and transmission grids, such as overvoltage and reverse power flow. Embedding voltage support schemes into PVs, such as standard ${\cos \phi (P)}$ characteristic proposed by the German grid codes, may cause more changes in the steady-state behavior of distribution grids and, in turn, the transmission side. Accordingly, it is important to properly model active distribution grids to analyze the system impacts of these changes to plan and operate future smart power grids. However, due to the high dimension of distribution grids, considering a detailed distribution grid to study the transmission side or a fraction of the distribution grid is either cumbersome or impractical. Therefore, it is required to develop a reasonable equivalent that can fairly capture the dominant behavior of the distribution grids. The aim of this paper is to use gray-box modeling concepts to develop a static equivalent of distribution grids comprising a large number of PV systems embedded with voltage support schemes. In the proposed model, the PV systems are aggregated as a separate entity, and not as a negative load, which is traditionally done. The results demonstrate the superior quality of the proposed model compared with the model with PV systems as the negative load.

This paper presents a technical evaluation of two different methods to avoid the excessive voltage deviations caused by dispersed generation in distribution grids. Two voltage regulation methods, wide area voltage control of HV/MV transformer and inverters’ reactive power support, are simulated in several real rural grids with different topologies located in Northern Germany. In all these analysed grids the high penetration of dispersed generation produces big power flow reversals during low load periods. The objective of implementing both voltage regulation methods is to reduce voltage deviations in such an extent that voltage band violations no longer act as the principal obstacle to host more dispersed generation in the distribution grid. We evaluate the effectiveness of these two methods through a power system analysis program and analyse the reciprocal effects caused by simultaneous use of both methods. Moreover, the concept of “voltage inhomogeneity” is introduced.

Evolutionary multitasking fuzzy cognitive map learning

Distribution networks (DNs) are facing numerous challenges such as variability of demands, environmental issues, high power losses, and fluctuating voltage profiles. Distributed energy resources (DERs) are becoming more important due to their economic and environmental impacts. This paper presents optimal siting and sizing of the Photovoltaics (PVs) and Wind Turbines (WTs) to improve the voltage magnitude profiles. This planning problem is formulated using DER generation and load profiles of the three representative days, one for each season. The resulting constrained optimization problem is solved using Cuckoo Search Algorithm (CSA). The proposed solution approach is applied to the 33 bus and 69 bus radial distribution networks. Several simulations are performed for the performance analysis of the methods and the results are compared to several available ones.