Control of wind power plants operation based on synchronized measurements

Intermitirajuća narav električne energije proizvedene iz vjetroelektrana čini velike probleme prilikom vođenja elektroenergetskog sustava i time ograničava njihovu veću integraciju. Paralelno s tim procesom, proces starenja postojeće infrastrukture također predstavlja dodatne izazove vlasnicima i operaterima infrastrukture zbog potreba za velikim kapitalnim investicijama. Korištenje naprednih i sinkroniziranih mjernih tehnologija nameće se kao jedno od rješenja za veću integraciju vjetroelektrana i obnovljivih izvora u cjelini. U ovom se radu opisuje rješenje vođenja pogona vjetroelektrane korištenjem tehnika rudarenja podataka na skupu sinkroniziranih mjernih podataka iz PMU jedinica i ostalih naprednih mjerenja. Pri tom se postojeća infrastruktura može koristiti kao osnovna platforma u kojoj se sve dodatne instalacije mogu brzo i jednostavno izvršiti modularno i adaptivno. U radu je opisan model upravljanja pogonom vjetroelektrana zasnovan na metodi klasterizacije te klasifikacije pogonskih stanja uporabom klasifikacijskih i regresijskih stabala odlučivanja te uporabom sinkroniziranih podataka generiranih iz mjernih i senzorskih uređaja. Algoritam odlučivanja o režimu pogona vjetroelektrane izrađen je u obliku stabla odlučivanja. Konačni model upravljanja i algoritam odlučivanja verificirani su korištenjem testnog modela te na mjernim podacima iz elektroenergetskog sustava. Pokazane su mogućnosti primjene ovakvih metoda i algoritama kao i proširivost na veće količine električkih i neelektričkih podataka koje će se generirati u elektroenergetskim sustavima.The intermittent nature of electricity produced from wind power plants poses major problems in the control of the power system and thus limits their bigger integration. Parallel to this process, the aging process of the existing infrastructure also poses additional challenges to infrastructure owners and operators due to the need for large capital investments. The use of advanced and synchronized measurement technologies is one of the solutions for greater integration of wind power and renewable resources as a whole. This thesis describes the solution of wind power plant operation using data mining techniques at a set of synchronized measurement data from PMU units and other advanced measurements. In this, the existing infrastructure can be used as a basic platform in which all additional installations can be done modularly and adaptively quickly and easily. This thesis describes a model of wind power plant control based on the clustering method and the classification of operating conditions by using classification and regression trees and using synchronized data generated from measuring and sensor devices. An increase in the number of intelligent electronic devices, ie various measuring, monitoring, control and sensing devices installed in the power system, results in a constant increase in the amount of data transmitted and stored in the competent control centers and other hierarchical levels. The amount of data described will be increasing in the future as a further increase in the number of installed subject devices can be expected. This raises the possibility and need to analyze these new types of data not only by classical deterministic models, but also by new non-deterministic mathematical models. One of the key prerequisites for this is data consistency, harmonization and synchronization. Having all the data concentrated would enable efficient data storing and processing enhancing the current information stream with the extraction of right information from the big data surrounding. Key characteristic of such future infrastructure enhancements would need to be adaptivity towards existing power grid infrastructure and modularity to allow system’s components separation and recombination (“adaptidular” infrastructure). The most important benefits of the new infrastructure (Figure 3) following the adaptidular design paradigm can be described as following: • Existing capacities and possibilities of existing infrastructure enhancement and upgrading • Capital expenditures (CAPEX) postponing or abolishing (building of new lines, substations, power infrastructure reconstruction) due to availability of new information in asset management systems, dynamic line rating system, PMU systems etc. • Maintenance cost cutting through the usage of predictive maintenance enabled through sensor networks and IoT gateways • Additional services provision: numerous additional services such as meteorological data assessment, air quality mapping, telecom services provision through IP/MPLS etc. Having all the data concentrated would enable efficient data storing and processing enhancing the current information stream with the extraction of right information from the big data surrounding. Key characteristic of such future infrastructure enhancements would need to be adaptivity towards existing power grid infrastructure and modularity to allow system’s components separation and recombination (“adaptidular” infrastructure). The most important benefits of the new infrastructure following the adaptidular design paradigm can be described as following: - Existing capacities and possibilities of existing infrastructure enhancement and upgrading - Capital expenditures (CAPEX) postponing or abolishing (building of new lines, substations, power infrastructure reconstruction) due to availability of new information in asset management systems, dynamic line rating system, PMU systems etc. - Maintenance cost cutting This thesis describes one of the possibilities of using such harmonized and synchronized data using a data mining technique for the purpose of operating a wind farm. In order to investigate the possibility and potential of wind power monitoring and control based on big data surrounding an algorithm for monitoring and incorporating synchrophasor measurement was developed. As described earlier, it has all the characteristics of adaptive and modular applications that can easily be installed and commissioned on the existing infrastructure. It also provides ability for later upgrades and integration into large scale applications. The power system infrastructure produces huge amounts of data. The nonlinear nature of this data makes the extraction of useful information complicated. Compared to standard mathematical models, data mining techniques are non-deterministic and provide a feasible and valid solution which is not exact but is simple to obtain, concise, practical and easy to understand. This characteristic is especially suitable when processing the big data streams which are inevitably involved. As mentioned earlier, large wind power capacities are being installed and connected to different voltage levels. Every wind turbine, wind measuring masts inside the wind park transformer substations, etc. represent the source of large quantities of data every second. All these data streams can be further expanded with the installation of new data sensors arrays. These large quantities of data can be deemed unnecessary, but with the usage of different big data algorithms a way to monetize this data can be found. The most important data that can and should be used in power system data mining algorithms is the data for state estimation and future power system state predictions. These data streams can be classified into three main groups: 1. Phasor values measurements; 2. Loads and production measurements; 3. Other influential variables measurements. Phasor values like voltages and currents together with belonging phasor angles, can be gathered through PMU measurements and can provide valuable insights into system operation. Also, load and generation data with exact time stamp can easily be measured and collected to afterwards be used for different analyses. Other influential variables of additional data that are not directly connected to power system monitoring and control are also sometimes highly influential. These include meteorological data from various kinds of measurement systems of which most important are wind speeds and wind directions, air temperature, humidity and pressure, solar irradiance measurements. Together with meteorological data, other measurements such as conductor temperatures, overhead line sags, partial discharges, current transmission line capacity obtained by dynamic line rating (DLR) systems etc. can also be collected. All these data series can be used in wind and solar power system monitoring and control as well as for load forecasting applications and power evacuation possibilities. The prerequisite is to have an efficient solution for data transmission and processing. As described earlier, the huge amounts of data inside power creates the big data surroundings. The non-linear nature of the system makes the definition of new models for extraction of useful information from heaps of gathered data even more demanding. Especially demanding is the usage of data from wind power plants since these stochastic sources produce even bigger amounts of data due to dependable variables which influence the output power. Therefore, good data mining scope thus integrates wide area of variables. This paper defines simplified model which comprises of: - Wind power plant active and reactive power production (PWind, QWind), at wind power plant point of common coupling (PCC); - Wind power plant active and reactive power settings (PSettings, QSettings), which are operational decisions for the settings of wind power controller placed at wind power (PCC); - Total system load measurements (PL), expressed in percentage, as a percentage of nominal load; - Voltage amplitudes and angles (phasors) measurements (Vi , δi ) on selected nodes in the system; - Line, transformer and generator availability information Each operating condition (OC) is defined as a mathematical set whose members are the following elements or variables: OCk = {V1, V2, V3, . . . Vi , δ1, δ2, δ3, . . . δi, PL, PWind, QWind, Zth} (1) - with i = 1, 2, 3, . . . n; where n is the number of nodes in power system with measurements of effective values and voltage angles in the system, and - with k = 1, 2, 3, . . . m; where m—total number of input states over which data mining techniques are analyzed. The abovementioned data can be expanded by defining the finely tuned fractal structures attached to it: • Wind power total can be divided into wind power of single wind turbine or a cluster of turbines; • Total system load can be divided into loads on busbar, consumer, or load area level; • Voltage amplitudes and angles can be enhanced with current amplitudes and angles for each branch as well as Thevenin impedance measurements; • Wind production is defined with wind speed and can further be detailed with wind direction, air temperature and pressure, solar irradiance and air humidity measurements; • Line and transformer availability can further be described through breaker status in line bays and transformer bays or through transformer and line monitoring systems. All this data needs to form large and well-organized databases for further usage in control, planning, asset management and operation and maintenance (O&M) optimization process. Therefore, to take full advantage of the available data efficient algorithms for big data analysis are needed. The proposed model of wind power plant management consists of a clustering part, which divides the existing metering and simulation data into three data clusters for normal, emergency and critical operating conditions, and then a classification part, thereby assigning previously defined data classes to the new metering data. The first scientific contribution of this paper was defined by the creation of a model of wind power plant control based on synchronized measurements. In this way, the new metrics can define the transition of the power system from normal to emergency or critical state of the system and define a new so-called. Early Warning Signal (EWS) as a measure of security awareness (SA - Situational Awareness). The aim of the developed algorithm is to create a new kind of early warning signal (EWS) and recognize the structure of critical transitions for transmission system and wind power operators in the form of a situational awareness (SA) indicator. These signals should be structured to warn the operators that the alarming operating condition could be reached and that preventive or corrective actions should be done (e.g., wind power curtailment or reactive power support increase) and thus move the system to normal operating state, like described in figure below. Created EWS signal as a situational awareness indicator serves as a main triggering signal for operating decisions in wind power settings in order to change operating condition back to EWS value NORMAL. Therefore, EWS could serve as a first line of defense to reduce the risks of total or partial system blackouts and thus reducing the opportunity costs associated with the costs of electric energy not being delivered. Commonly used data mining algorithms identified by the IEEE International Conferences on Data Mining (ICDM) are C4.5, k-Means, Support Vector Machine (SVM), Apriori, PageRank, AdaBoost, Neural Networks, Naive Bayes and Classification and regression trees (C&RT). These 10 algorithms cover classification, clustering, statistical learning, association analysis, and link mining, which are all among the most important topics in data mining research and development. The first step in the algorithm is data management and preparation which consists of time synchronization, format unifying, and ordering of historical raw data from actual power system measurements. Additionally, synthetic data which is produced and gathered from various kinds of simulations based on mathematical models is also included in this step. In this way, mathematically defined power system states are defined as input data in the algorithm. It is important to note that except for the variables defined herein, the input set of system states can be extended to a whole range of additional input signals such as data from various measuring devices for measuring electrical and nonelectric values, meteorological measuring devices, sensors and other devices. The model is therefore adaptive and modular. It is easy to upgrade by simply expanding the operating condition (OC) math data set. The second step is data clustering, with the aim of defining system states on a given database or set of operating conditions. For the algorithm design described in this paper, the analytics software package Statistica was used. Standard variable definition from statistical theory was used where an independent variable (also called experimental or predictor variable), is being manipulated in an experiment to observe the effect on a dependent variable (also called an outcome variable). Total set of operating conditions in this example to be a representative sample needs to be large enough and cover all possible system states and. K-Means algorithm with Euclidian distances was used for clustering of the initial data set. The classification part of the model results in a decision tree that contains measurement variables, such as synchronized measurements, and control variables, such as wind power plant active and reactive power. The induced decision tree results in possible operating conditions that belong into normal, emergency and critical operating states clusters. As a final part of the operating mode decision algorithm, a simple optimization procedure is performed by selecting a possible normal operating condition with minimal reactive power production and maximum active power production. By defining the decision tree described above and performing a simple optimization procedure, a decision algorithm for changing the mode of operation of wind power plants was defined based on the characteristic values of the electrical parameters of the system in order to optimize the power system operation, and thus the second scientific contribution of this paper was defined. Verification of the developed mathematical model based on measurements in a real power system was made using measurements made in the actual operation of the Zadar 6 wind power plant, connected to the 35 kV distribution network, and the wind power plants Zadar 2 and Zadar 3, connected to the 110 kV transmission network. Verification of the mathematical model confirmed the tested mathematical model and the algorithm developed, and the third scientific contribution of this paper was defined. The developed and tested model can be considered robust, since any type of parameter measurements identified as essential parameters for the operation of wind farms and the power system as a whole can serve as input to the algorithm. An essential prerequisite is that all metrics are standardized, harmonized and synchronized, which is quite difficult to achieve at this time, especially given the non-harmonization of measurement and communication standards and protocols for these different parameters. Also, a necessary prerequisite is the existence of a record of operating states in all defined operating areas, ie normal, emergency and critical operating conditions. It is very difficult to obtain records of a state of emergency, and especially critical, operating condition in a real facility. Therefore, records in critical operating condition need to be made most often by simulations or so called synthetic or syntex data. Continued research in this area primarily involves the production of more simulation records and data in the emergency and critical drive section. This gives a more granular granulation when creating a data cluster. Further upgrades include taking into account additional sets of other types of metrics, such as sensor data, in particular, from a series of sensors that will increasingly be installed in the power system. Concerning the application of the mathematical model and the algorithm developed, it is currently necessary to define the market in the Croatian Power System with a reactive power regulation service. It is also necessary to systematically define ways of limiting the wind farm's operating power in terms of covering losses of wind farm owners