One-minute data

Overview

Accurate photovoltaic (PV) power plant design requires a comprehensive analysis of high-resolution solar resource data during the initial design phase. While traditional approaches often rely on hourly or 15-minute data intervals, the increasing complexity of modern PV systems demands more precise temporal resolution. One-minute resolution data provides critical insights into several key aspects that significantly impact PV power output (PVOUT) estimation:

• System configuration optimization: The relationship between DC array sizing and AC inverter capacity directly affects annual energy yield. One-minute resolution data enables precise analysis of power conversion dynamics, helping designers optimize the DC/AC ratio while minimizing production losses.

• Power output variability: Short-term fluctuations in solar resource, particularly in regions with variable cloud cover, can significantly impact power generation. These rapid changes, often masked in lower temporal resolution data, affect both system performance and grid integration requirements.

• Grid integration requirements: Modern grid codes demand sophisticated control of power quality parameters. Understanding rapid variations in PV output helps designers implement appropriate technical solutions for grid compliance and stability maintenance.

• Site-specific characteristics: Local meteorological conditions and their impact on PV performance can only be fully assessed using high-resolution data. This enables a more accurate evaluation of site suitability and expected system performance.

System configuration optimization

Photovoltaic system design requires careful optimization of the DC/AC ratio - the relationship between PV module power at Standard Test Conditions (STC) and the inverter's AC power rating. This ratio directly impacts system performance, clipping losses, and financial outcomes.

The DC/AC ratio optimization represents a multi-criteria problem that aims to minimize investment and operating costs while maintaining compliance with technical regulations. When the DC/AC ratio is too high, the PV array may produce more power than the inverter can handle, resulting in clipping losses that impact overall system efficiency.

Key considerations for DC/AC ratio optimization encompass several interrelated factors. The PV module technology and mounting configuration significantly influence real-world performance compared to STC conditions, while inverter efficiency characteristics and power rating limitations determine conversion capabilities. Geographic location factors also play a role, especially in areas experiencing frequent cloud edge effects that can impact power output. Additionally, site-specific radiation levels and their variability throughout the year must be carefully evaluated, along with temperature conditions that affect module output efficiency. These factors must be considered to achieve optimal system design and performance.

One-minute resolution data provides significantly more accurate insights compared to 15-minute or 60-minute intervals:

• Power output variations are more precisely captured, revealing rapid fluctuations masked by longer averaging periods (fig.1).

• PV simulations typically underestimate clipping losses when using 15-minute or 60-minute data, particularly with larger DC/AC ratios.

• Annual clipping losses vary significantly with the DC/AC ratio, reaching 2-5% at a ratio of 1.5.

Fig.1: Simulated PVout during a specific day: calculation based on 1-min, 15-min, and 60-min input data

Power output variability

Solar power plant output fluctuations significantly impact system performance and grid integration requirements. Understanding these variations through high-resolution data analysis is crucial for accurate system design and operation.

One-minute resolution data reveals critical power output variations that are often masked in longer averaging periods:

• Short-term irradiance changes due to cloud edge effects create rapid fluctuations in power output.

• Sites with frequent intermittent clouds show particularly significant variations.

• Power fluctuations are typically underestimated when using 15-minute or 60-minute data intervals.

The accurate assessment of power output variability through 1-minute resolution data enables better system design decisions and more precise production estimates, particularly in regions with variable weather conditions.

Grid integration requirements

High-resolution data analysis ensures proper grid integration of photovoltaic systems, particularly regarding power quality and grid compliance requirements.

One-minute resolution data provides essential insights for grid integration:

• Accurately captures rapid PV output fluctuations masked by longer averaging periods.

• Enables precise assessment of power quality parameters.

• Helps determine appropriate equipment specifications for grid compliance.

• Facilitates better planning of ancillary services.

The use of 1-minute resolution data enables more precise grid integration planning and helps ensure reliable supply of quality electricity to consumers while maintaining compliance with grid operator requirements.

Site-specific characteristics

Site location and local environmental conditions significantly impact PV system performance and the accuracy of power output predictions when using different temporal resolutions.

One-minute resolution data analysis across climate regions reveals distinct patterns in how temporal resolution affects power predictions:

• Tropical climates show the highest percentage differences between 1-minute and longer interval calculations.

• Dry climates exhibit the lowest differences despite having the highest overall GTI values.

• Temperate and continental climates show moderate differences.

The impact of temporal resolution varies significantly based on site-specific characteristics, making individual site analysis crucial for accurate system design and performance prediction.

Temporal resolution impact analysis

The following study analyzed 20 sites across different climate groups to demonstrate the impact of temporal resolution on PV output calculations. Five representative sites were selected for detailed analysis, covering continental climate with the lowest GTI, dry climate with the highest GTI, tropical climate with the highest D2G ratio, dry climate, and tropical climate with the highest PVOUT difference.

The analysis used three temporal resolutions: 1-minute data (highest resolution), 15-minute data, and 60-minute data. DC/AC ratios of 1.1, 1.3, and 1.5 were evaluated to assess clipping losses under different system configurations.

Analysis of the sites revealed:

• The site with the dry climate with the highest GTI showed the highest annual clipping losses but the lowest percentage difference between temporal resolutions.

• The site with the tropical climate with the highest D2G ratio demonstrated the third lowest annual clipping losses but the highest difference between temporal resolutions.

• The site with the tropical climate with the highest PVOUT difference exhibited significant PVOUT overestimation, reaching over 2% at a DC/AC ratio of 1.5 using 60-minute data.

• Tropical sites show up to 2.5% overestimation of annual production when using 60-minute data with a DC/AC ratio of 1.5 (fig.3).

• Sites with high D2G ratios require particular attention to short-term power output variability effects.

• Sites with frequent intermittent clouds show significant power fluctuations.

• Power quality control measures can be more accurately specified using 1-minute resolution data.

• Sites with moderate PV output and GTI variability show the highest percentage inaccuracy in production calculations when using 60-minute resolution data compared to 1-minute resolution data.

A detailed example from one site shows how PVOUT varies across temporal resolutions during a single day with inverter clipping (fig.2). The 60-minute data (green line) shows no clipping, leading to overestimation compared to 1-minute and 15-minute measurements.

Fig.2: PVOUT at site with the tropical climate with the highest PVOUT difference, DC to AC ratio 1.3: a day with inverter clipping occurring with 1-minute and 15-minute data.

Fig.3: Overestimation of annual PVOUT for different time resolutions and DC to AC ratio.

Conclusion

The analysis of 1-minute resolution data in photovoltaic system design reveals several critical advantages for achieving optimal system performance across different climate zones and system configurations.

The DC/AC ratio optimization becomes more precise with 1-minute data, enabling more accurate sizing decisions that balance investment costs with performance. Annual energy calculations using 15-minute or 60-minute data can overestimate production by several percent compared to 1-minute assessments, with clipping losses reaching 2-5% at DC/AC ratios of 1.5.

Different climate zones demonstrate distinct patterns in how temporal resolution affects power predictions. Tropical climates show the highest percentage differences between 1-minute and longer interval calculations, while dry climates exhibit the lowest differences despite having the highest overall GTI values. Sites with moderate PV output and GTI variability demonstrate the highest percentage inaccuracy in production calculations.

One-minute resolution data enables better management of voltage variations and fluctuations, harmonic distortion, power frequency variations, and energy storage system sizing for output smoothing. This granular data proves particularly valuable for grid compliance and system stability.

Adopting 1-minute resolution data represents a significant advancement in PV system design methodology. It provides more accurate performance predictions and enables better-optimized installations across diverse geographical locations and climate conditions. Using 15-minute data instead of 60-minute data typically reduces production overestimation by approximately 50%, though significant differences from 1-minute data remain.

Further reading

• “Impact of time resolution of solar and meteorological data on clipping losses and energy yield simulation” by J.Rusnak, B. Schnierer, M. Suri, M. Opatovsky, G. Srinivasan.