Kristian Balzer

Dr. Kristian Balzer

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 January 2022 in IEEE Latin America Transactions
Verified
Journal Article
Closed access
KBKristian Balzer
Kristian Balzer
DWDavid Watts
David Watts
In Chile, the operation and market of renewable solar-wind generation and conventional gas-coal-diesel-hydraulic generation is made up of two electric generation systems that are managed in a complementary manner and currently interconnected under the name of Sistema Electrico Nacional, counting on the greatest potential and solar development in South America. However, the real-time operation is carried out by the Dispatch and Control Center, which treats them as two independent electrical generation systems in terms of supervision and control. The system faces the challenge associated with the massive potential and development of solar-wind generation concentrated in the northern area of Chile. This creates difficulties in its technical operation, especially in frequency control and the complexity of modeling in real-time the changing market for complementary services. Therefore, the primary frequency control is the complementary service that acts as the first action and line of defense to allow the system to integrate large blocks of renewable energy. It is necessary to deliver a minimum of dynamic reserves in times of ten seconds and five minutes to avoid the operation of stages of load disconnection / generation due to low / over frequency in front of large power disconnections of gas-fired plants (520 MW), hydraulic of reservoir (700 MW) and coal (380 MW) or loss of demand from large mining processes (300, 500 and 600 MW) for systemic frequency variations that should not exceed 0.7 Hz.

29 February 2024 in Applied System Innovation
Verified
Journal Article
Open access
PDF file
KBKristian Balzer
Kristian Balzer
DWDavid Watts
David Watts
Modern electrical power systems integrate renewable generation, with solar generation being one of the pioneers worldwide. In Latin America, the greatest potential and development of solar generation is found in Chile through the National Electric System. However, its energy matrix faces a crisis of drought and reduction of emissions that limits hydroelectric generation and involves the definitive withdrawal of coal generation. The dispatch of these plants is carried out by the system operator, who uses a simplified mechanism, called “economic merit list” and which does not reflect the real costs of the plants to the damage of the operating and marginal cost of the system. This inefficient dispatch scheme fails to optimize the availability of stored gas and its use over time. Therefore, a real-time redispatch model is proposed that minimizes the operation cost function of the power plants, integrating the variable generation cost as a polynomial function of the net specific fuel consumption, adding gas volume stock restrictions and water reservoirs. In addition, the redispatch model uses an innovative “maximum dispatch power” restriction, which depends on the demand associated with the automatic load disconnection scheme due to low frequency. Finally, by testing real simulation cases, the redispatch model manages to optimize the operation and dispatch costs of power plants, allowing the technical barriers of the market to be broken down with the aim of integrating ancillary services in the short term, using the power reserves in primary (PFC), secondary (SCF), and tertiary (TCF) frequency control.

30 November 2023 in Preprints.org
Verified
Preprint
Open access
PDF file
KBKristian Balzer
Kristian Balzer
DWDavid Watts
David Watts
Modern electrical power systems integrate renewable generation, with solar generation being one of the pioneers worldwide. In Latin America, the greatest potential and development of solar generation is found in Chile through the National Electric System. However, its energy matrix faces a crisis of drought and reduction of emissions that limits hydroelectric generation and involves the definitive withdrawal of coal generation. The dispatch of these plants is carried out by the system operator, who uses a simplified mechanism, called “economic merit list” and which does not reflect the real costs of the plants to the damage of the operating and marginal cost of the system. This inefficient dispatch scheme fails to optimize the availability of stored gas and its use over time. Therefore, a real-time redispatch model is proposed that minimizes the operation cost function of the power plants, integrating the variable generation cost as a polynomial function of the net specific fuel consumption, adding gas volume stock restrictions and water reservoirs. In addition, the redispatch model uses an innovative "maximum dispatch power" restriction, which depends on the demand associated with the automatic load disconnection scheme due to low frequency. Finally, by testing real simulation cases, the redispatch model manages to optimize the operation and dispatch costs of power plants, allowing the technical barriers of the market to be broken down with the aim of integrating ancillary services in the short term, using the power reserves in primary (PFC), secondary (SCF), and tertiary (TCF) frequency control.

3 September 2023 in Energies
Verified
Journal Article
Open access
PDF file
KBKristian Balzer
Kristian Balzer
JLJoaquin Lazo
Joaquin Lazo
DWDavid Watts
David Watts
Modern power systems integrate ancillary services (ASs) to provide security and quality of service in real-time operation (RTO) due to the intense frequency variations caused by the uncertainty of solar–wind generation. To this end, the ancillary services market focuses on power reserves for secondary and tertiary frequency control. The adjustment and dispatch of reserves from plants are manual instructions executed by the system operator in order to maintain the frequency within the normal operating range (49.80 ≤ f ≤ 50.20 Hz). This work proposes a methodology for the economic modeling of the ancillary services market in real time using a dynamic hourly mathematical model that integrates the variability of solar–wind generation, a demand monitoring curve, and the trajectory of marginal cost (MgC). This is a segmented methodology in which plants with costs close to the marginal cost are identified in real time using the Supramarginal (SMg) and Inframarginal (IMg) methods. Finally, this economic model for real-time power reserve reallocation represents an innovation in the ancillary services market because its results allow for the operation costs (OC) of the reserves to be reduced by up to 60% and for the displacement of marginal costs to be reduced by 10 to 40% with respect to traditional methodologies such as the economic merit list and the technical minimum methodology, which cause plants to operate without economic justification.

8 August 2023 in Preprints.org
Verified
Preprint
Open access
PDF file
KBKristian Balzer
Kristian Balzer
JLJoaquin Lazo
Joaquin Lazo
DWDavid Watts
David Watts
Modern electrical power systems integrate ancillary services to provide security and quality of service in real-time operation because of the intense variations in frequency caused by the massive development and uncertainty of solar-wind generation. Therefore, this ancillary services market focuses on power reserves for secondary and tertiary frequency control. Adjusting reserves and dispatching plants is a manual instruction executed by the system operator to maintain the frequency in the normal operating state (49.80≤f≤50.20 Hz). However, in the absence of an economic model for real-time power reserve reallocation in the ancillary services market, the reserve adjustments made by the system operator are not always optimal since they generate a displacement between the scheduled and actual marginal costs. Then, this work proposes a methodology for operating the ancillary services market in real-time through a dynamic and hourly mathematical model that integrates the variability of solar-wind generation, the demand monitoring curve, and the trajectory of the marginal cost. This model minimizes power reserve costs, which are governed by hourly price auctions, for candidate plants classified as supra/infra-marginal and can optimally reallocate power reserves for secondary and tertiary frequency control.

11 July 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
BVBárbaro Maykel López-Portilla Vigil
Bárbaro Maykel López-Portilla Vigil
KBKristian Balzer
Kristian Balzer
LCLorena Carballo
Lorena Carballo
MFMiguel Figueroa
Miguel Figueroa
FERNANDA PESETFERNANDA PESET
FERNANDA PESET
Pedro Fernández de Córdoba CastelláPedro Fernández de Córdoba Castellá
Pedro Fernández de Córdoba Castellá
Detecting defective pixels in CMOS Image Sensors is a critical task for ensuring high-quality image acquisition, as even minor defects can significantly degrade performance in vision systems and downstream processing. This study evaluates the performance of several supervised machine learning algorithms for defective pixel detection using 300 grayscale images of 512×512 pixels from the publicly available TAMPERE17 dataset, in which dead and hot pixels were randomly introduced. The evaluated algorithms include k-Nearest Neighbors (kNN), Naïve Bayes, Decision Tree, Random Forest, Support Vector Machine (SVM), and Multilayer Perceptron. Performance was assessed using specificity, recall, precision, and the phi coefficient. A sensitivity analysis was additionally performed using neighborhood windows of 3×3, 5×5, and 7×7 pixels during feature extraction. The results showed that the neighborhood size has a noticeable impact on classification performance, with the best overall results obtained using an SVM classifier with an RBF kernel and a 3×3 feature extraction window. This configuration achieved a phi coefficient of 98.08%, together with a specificity of 99.00%, a recall of 99.00%, and a precision of 99.02%. Statistical analysis further indicated that SVM (RBF) and Multilayer Perceptron exhibited statistically comparable performance under the evaluated experimental conditions. The analysis further revealed that defective pixels are primarily characterized by highly local intensity variations, while larger neighborhoods do not necessarily improve classification performance. Compared with previously reported defective pixel detection methods, the proposed approach achieved highly competitive results and the highest phi coefficient among the evaluated methods. Additionally, the impact of defective pixel detection on image quality was assessed using a simple median filter correction stage and evaluated through Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). The best-performing configuration achieved a PSNR of 40.96 dB and an SSIM of 98.68% after median filter correction, demonstrating substantial improvement over median filtering alone, although the resulting image quality remains below that of the best-performing state-of-the-art correction methods. The proposed framework provides reproducible reference results for evaluating supervised machine learning approaches to defective pixel detection while analyzing the influence of feature extraction window size on classification performance and reconstructed image quality.