5 Ways to Use Remote Monitoring to Increase Power Generation | Early Detection of Anomalies

Table of Contents

• Remote monitoring to increase power generation begins with early detection of anomalies

• Usage 1: Check daily and time-of-day power generation curves

• Usage 2: Compare output differences at the PCS and string unit levels

• Usage 3: Cross-check alarm and shutdown histories with the timing of power generation declines

• Usage 4: Separate weather and seasonal variations from site-specific factors

• Usage 5: Link remote monitoring results to on-site inspections and records

• Points to note to avoid making judgments based solely on remote monitoring

• Summary

Remote monitoring to increase power generation begins with early detection of anomalies

When you want to increase the power generation of a solar power system, it is important to review not only on-site cleaning and inspections but also how you use remote monitoring. Remote monitoring is a system that lets you grasp power generation, the status of the PCS, stoppage history, alarms, and generation curves by time of day without going to the plant. However, simply looking at the screen will not lead to improved generation. What matters is using remote monitoring not as a tool to check whether an anomaly has occurred, but as a tool to quickly detect generation losses and determine the priority of on-site inspections.

What operations personnel searching "how to increase power generation" want to know are practical judgments: what to check when generation is low, which anomalies to prioritize, and how far they can isolate the problem before going to the site. By leveraging remote monitoring, it becomes easier to quickly determine whether a drop in generation is due to temporary weather, a fault in specific equipment, a PCS stoppage, or a decline at the string level. Detecting anomalies early reduces wasted cleaning, weeding, repairs, and equipment checks, and enables responding starting from the locations that have the greatest impact on power generation.

In solar power generation, you cannot increase the amount of solar irradiance itself at the site. You cannot increase the number of sunny days, nor can you change the sun's altitude with the seasons. However, you can move closer to a state in which the received irradiance is converted into electricity with as little waste as possible. The role of remote monitoring is to quickly detect these "missed generations." By capturing signs such as output not rising during hours when generation should be possible; a section being lower compared with equipment under the same conditions; an unnatural generation curve on a clear day; or repeated short stoppages during the daytime, the causes of generation decline can be isolated early.

There are anomalies detected by remote monitoring that are obvious and others that are less obvious. If the PCS is completely stopped it is easy to notice, but brief stoppages, output plateauing, reduced performance in some strings, shading only in the morning and evening, and gradual declines caused by soiling at the lower edge can be missed if you look only at monthly generation figures. To avoid losses in long-term operation, you need to use remote monitoring routinely and confirm small changes in generation as early as possible.

However, the cause cannot be determined by remote monitoring alone. Causes of low power generation are diverse, including dirt on the panel surface, shadows from weeds or trees, faults in connection points, cable damage, PCS shutdowns, output curtailment, temperature rises, and poor drainage. Remote monitoring is not meant to identify the cause; it is an entry point to narrow down what to look for on site. By detecting anomalies in the data, cross-referencing them with photos and location information on site, and confirming changes in power generation after countermeasures, the accuracy of power output improvements increases.

How to use 1: Check daily and time-of-day power generation curves

The basics of remote monitoring are to look not only at daily power generation but also at the time-of-day generation curve. If you only look at monthly or cumulative generation, you can't tell when generation losses are occurring. Likewise, even when overall generation is low, the suspected causes differ depending on whether it's low only in the morning, the midday peak fails to develop, it falls early in the evening, or there are sudden dips during the day. To increase generation, it is important to first read the direction of anomalies from the shape of the generation curve.

If the morning ramp-up is slow, suspect shadows on the east or southeast side. Shadows from weeds, trees, slopes, fences, surrounding structures, or adjacent equipment may be reducing morning power generation. If evening power drops off early, check for shadows on the west or southwest side, surrounding terrain, and tree growth. If the midday peak does not reach expected levels, candidates include soiling on the panel surface, temperature rise, PCS output limiting, output curtailment, or insufficient input. If the generation curve suddenly falls partway through, correlate PCS stop history and alarm history with the timestamps.

In remote monitoring, it is important to use the power generation curve on clear-sky days as a reference. On cloudy or rainy days, power output fluctuates widely due to cloud movement, making it difficult to determine whether a change is caused by equipment malfunction or by the weather. On clear-sky days, it is easier to spot recurring shadows that fall at the same times, output clipping that flattens the tops, valleys caused by sudden stops, and persistent declines in specific equipment. Rather than looking only at days with low generation, it is necessary to compare with days of similar weather and with past clear-sky days.

When looking at a power generation curve, consider steep drops and gradual declines separately. Steep drops may be related to PCS shutdowns, alarms, poor connections, or the occurrence of obstructions. Gradual declines may be related to dirt on panel surfaces, the growth of weeds or trees, site environment deterioration due to poor drainage, or aging of equipment. Using remote monitoring, you can check for these types of declines before going to the site.

In daily checks, the important thing is that when you find signs of an anomaly, you follow up with an on-site inspection. If readings are low only in the morning, check for morning shadows; if low only in the evening, check for evening shadows. If the midday peak does not rise, check panel soiling, PCS history, temperature conditions, and output curtailment. Even if you find an anomaly on the remote monitoring screen, you cannot determine the correct countermeasure without checking on-site soiling, shadows, vegetation height, connection points, and the environment around the equipment. The power generation curve is the first piece of information for determining the direction of an on-site inspection.

Usage 2: Compare output differences at the PCS and string level

To leverage remote monitoring for improving power generation, it is important to look not only at plant-level figures but also at output differences at the PCS unit and string levels. If you only look at the plant’s total generation, some anomalies will be obscured by the average. Even if it doesn’t appear to be a major problem overall, there can be cases where only a particular PCS has low output, only a particular string remains consistently low, or part of the generation curve differs compared with equipment under the same conditions.

When comparing, it is fundamental to examine systems under the same conditions. Simply comparing systems that differ in orientation, tilt, number of panels, shading conditions, or connection configuration can cause you to mistake normal differences for abnormalities. The purpose of comparison is to identify locations that are consistently lower among systems that should inherently have similar power generation. If only part of a system is lower compared with adjacent rows or other systems with the same orientation, suspect soiling, shading, connection faults, cable damage, or an abnormality on the PCS side.

When output is low at the PCS unit level, check not only the PCS itself but also the input side. Even if the PCS output appears low, there may actually be insufficient power reaching the PCS due to input-side issues such as panel soiling, shading from weeds or trees, poor string connections, or cable damage. Conversely, if there are no major problems on the input side but the PCS generation curve is abnormal, check the PCS shutdown history, alarm history, any output curtailment, and the temperature environment.

String-level declines should be detected as early as possible. Even small differences in power generation can lead to large losses if they persist over a long period. If a string is consistently lower than its surroundings even on sunny days, possible causes include dirt, shading, poor connections, panel defects, and cable damage. If it is low only in the morning and evening, check for shading; if it becomes unstable after rain, suspect moisture or dampness around the connection points. Narrowing down the abnormal range with remote monitoring clarifies which locations should be checked during on-site inspections.

When you find an output discrepancy, it's important not to conclude there's a fault immediately. Remote monitoring only reveals differences in power generation or output. Whether that discrepancy is caused by soiling, shading, an electrical anomaly, or a limitation on the PCS side should be determined by combining on-site inspection and historical review. Recording the equipment ID suspected of being abnormal, the time of occurrence, changes in power generation, the comparison reference, and on-site photos will speed up future decisions.

Use Case 3: Correlate alarm and shutdown history with times of power output drops

One of the anomalies that can be detected early through remote monitoring is PCS shutdowns and alarms. However, simply checking whether an alarm is present is not sufficient. To increase power generation, you need to determine the extent to which alarms and shutdowns are contributing to decreases in generation. To do that, cross-check the time when the generation curve dips with the occurrence times of alarm and shutdown histories.

Even short stoppages can cause significant losses if they occur during the daytime when solar irradiance is strong. They may not be obvious in monthly generation totals, but on the generation curve they can appear as sharp dips during the day. If stoppages and restarts are occurring repeatedly during the day, the operational risk can be greater than the total downtime alone. Check whether the stoppage was a single occurrence, is recurring on the same PCS, or is happening simultaneously across multiple PCS.

If the alarm history coincides with a drop in power generation, prioritize checking the equipment. If the same alarm is repeatedly occurring on a specific PCS, check the PCS unit itself, the input side, the output side, and the surrounding environment. If it is occurring simultaneously across multiple PCS units, you need to check external conditions, common equipment, and output curtailment. Examining the weather, temperature, and site conditions at the time the alarm occurred makes it easier to narrow down potential causes.

On the other hand, there are cases where power generation has decreased but there are no alarms or shutdown logs. In such cases, check for panel surface contamination, shadows from weeds or trees, a drop in input-side string performance, output curtailment, temperature conditions, measurement anomalies, and so on. If you assume everything is normal simply because there are no alarms, you may overlook generation losses. In remote monitoring, it is important to look not only at the presence or absence of alarms but also at the shape of the generation curve and differences between equipment.

To make alarm and shutdown histories useful for on-site inspections, link the histories with location information and on-site photos. Verify which PCS experienced an anomaly, when it occurred, and what type of anomaly it was, and check the surrounding area for grass, sediment, standing water, or poor ventilation. If anomalies tend to occur after rain, check for moisture and drainage. If anomalies tend to occur during periods of high temperature, check ventilation and the temperature environment around the equipment. It is important to use remote monitoring logs to concretely determine what to look for on-site.

Usage 4: Distinguish weather/seasonal differences from site-specific factors

When using remote monitoring, declines in power output become easier to detect, but it is necessary to distinguish whether that decline is a natural result of the weather or a generation loss caused by on-site factors. Because solar power generation is heavily influenced by solar irradiance, output will drop during periods with many cloudy or rainy days even if the equipment has no problems. When remote monitoring identifies days with low output, do not immediately conclude there is a fault; first check the weather and seasonal variations.

If the entire power plant is showing a similar decline, the influence of weather and solar irradiance conditions may be significant. On the other hand, if only some PCS units or certain strings within the same plant are underperforming, and there is a clear difference compared with equipment under the same conditions, suspect an on-site cause. In remote monitoring, it is important to distinguish overall trends from localized anomalies. Even simply determining whether the decline is overall or localized will change the priority of on-site inspections.

Seasonal differences are also points that can be easily confirmed by remote monitoring. In spring, pollen, yellow dust, and airborne particulates can soil panels and affect power generation. From the rainy season through summer, weeds grow and shadows are more likely to occur in the morning and evening. In summer, panel temperatures and temperatures around equipment rise, which can make it harder for output to increase. After typhoons or heavy rain, check for fallen debris, sediment, poor drainage, and moisture around connection points. In autumn, fallen leaves, and in winter, shadows caused by the low sun angle, frost, and snow can have an impact.

In remote monitoring, it is useful to track seasonal changes in the generation curve. If only in winter the morning and evening ramp-up or ramp-down are poor, suspect shading caused by the lower solar altitude. If on sunny summer days the midday peak is muted, check the temperature environment and any output curtailment. If overall performance is poor in spring, check the cleanliness of the panel surfaces. By taking these seasonal characteristics into account, you can reduce misinterpretation of remote monitoring data.

By separating weather and seasonal variations from on-site factors, you can reduce unnecessary countermeasures. If the weather is the cause, cleaning or repairs will not significantly improve power generation. Conversely, if only a specific area remains low even on sunny days, there is a high likelihood that there is room for improvement on site. When using remote monitoring, it is important not only to detect a drop in power generation but also to determine whether that drop can be addressed.

How to Use 5: Linking Remote Monitoring Results to On-site Inspections and Records

Even if anomalies are detected by remote monitoring, unless they are linked to on-site inspections and records, it will be difficult to improve power generation. Remote monitoring is a convenient system that allows you to check power output and history without going to the site, but you cannot judge from the screen alone issues such as dirt on the panel surface, weed height, tree shadows, puddles around connection points, cable condition, or poor ventilation around equipment. To increase power output, you need a process that uses remote monitoring to narrow down the anomalous area, confirms the cause on site, and then verifies the power output again after countermeasures.

Remote monitoring allows you to prioritize on-site inspections. For example, if a system has low power generation only in the morning, check for shading in that direction. If only a specific PCS drops during the daytime, check shutdown history, alarm history, temperature conditions, and the input side. If only a particular string is continuously low, focus on soiling, shading, connection points, and cables. If you identify in advance the locations and times of reduced generation, you no longer need to conduct aimless inspections on site.

When conducting on-site inspections, it is important to record data using the same units as the remote monitoring information. Link which PCS, which string, which column, and which time period showed anomalies to on-site photos and location information. If the locations that showed reduced power generation in the data actually have dirt or shading, the rationale for countermeasures becomes clear. Conversely, if no dirt or shading is found on site, you can prioritize checking the connection points or the PCS itself.

After countermeasures are implemented, we confirm changes in power generation by remote monitoring. We check whether the midday peak improved after cleaning, whether the morning and evening drops improved after weeding, whether shutdowns or alarms decreased after improving ventilation around the PCS, and whether differences between strings became smaller after inspecting connection points. Measures that show an effect are prioritized in future work, and if the effect is small we suspect other causes. Without this verification of effectiveness, it is not possible to determine whether the work truly led to improvements in power generation.

Record keeping is also important. If you record the date and time an anomaly was detected by remote monitoring, the affected equipment, the characteristics of the generation curve, on-site photos, the work performed, and the power generation after countermeasures, you can make quicker judgments when the same anomaly recurs. Rather than searching for the cause from scratch each time, keeping past anomalies and countermeasure results available for reference makes it easier to reduce long-term operational generation losses. Remote monitoring is not something that functions in isolation; it becomes effective for improving power generation only when combined with on-site inspections and record management.

Precautions to Avoid Making Decisions Based Solely on Remote Monitoring

Remote monitoring is highly effective for early detection of abnormal power generation, but it is important not to determine the cause based solely on on-screen data. Although the result—low power generation—may be the same, the causes can vary widely, such as soiling, shading, weather, PCS shutdown, output curtailment, string faults, connection points, cables, temperature conditions, and poor drainage. Remote monitoring mainly reveals power output and changes in historical trends, and an on-site inspection is necessary to determine what is actually happening at the site.

For example, if the midday peak does not increase, dirty panels may be the cause, but output curtailment, temperature rise, or PCS limitations may also be involved. If generation is low only in the morning, shadows may be the cause, but dirt on the east side or input-side faults may also be related. If only a specific PCS is low, you need to check not only the PCS unit itself for abnormalities but also the input-side panels and strings, and surrounding shadows or dirt.

Remote monitoring data can appear differently depending on the measurement units and the display units. Looking only at total power generation can make some anomalies hard to see, and you may not be able to narrow down the cause without examining data at a finer granularity. Conversely, looking only at fine-grained units can lead to misinterpreting normal differences in conditions as anomalies. It is necessary to make comparisons taking into account orientation, tilt, number of panels, connection configuration, and surrounding shading conditions.

Also, the absence of an alarm does not necessarily mean the system is operating normally. Some generation losses do not trigger alarms. Dirt on the panel surface, shadows from weeds or trees, gradual output declines, and temperature effects due to poor ventilation may not manifest as clear alarms. When remote monitoring reveals a discrepancy in generation output, do not judge solely by the presence or absence of alarms; check the generation curve, equipment-to-equipment comparisons, seasonal variations, and on-site records together.

To use remote monitoring effectively, the accuracy of on-site records is also important. If you don’t know where cleaning was done, where weeding took place, which areas around the PCS were inspected, or where puddles were present, it becomes difficult to link remote monitoring data with on-site countermeasures. By linking remote monitoring with on-site records, you can establish a workflow for detecting anomalies, confirming causes, implementing countermeasures, and verifying their effectiveness.

Summary

What’s important in using remote monitoring to increase power generation is to examine generation data in detail, detect anomalies early, and link those findings to on-site verification. In solar power generation you cannot increase the solar irradiance itself at the site. However, you can improve output by bringing the system closer to a state in which the received irradiance is converted into electricity without waste. To do that, it is necessary to check, in sequence, the daily and time-of-day generation curves, output differences at the PCS and string levels, alarm and stoppage histories, weather and seasonal variations, and on-site inspection records.

In remote monitoring, we first check the power generation curve on sunny days to see whether output is low only in the morning, whether the midday peak fails to develop, whether it falls in the evening, or whether there are sudden dips during the day. Next, we compare PCS units and strings under the same conditions to see if any equipment is consistently underperforming. Correlating the time of the power drop with alarm and shutdown histories makes it easier to detect anomalies on the PCS side. However, even when no alarms are present, issues such as soiling, shading, poor ventilation, or faulty connections may still be present.

Also, it is important not to determine the cause based solely on remote monitoring. Once you narrow down the abnormal area using the data, check on site the panel surfaces, weeds, shadows from trees, connection points, the area around the PCS, drainage, and access paths. After performing cleaning, weeding, repairs, and equipment inspections, use remote monitoring to confirm how power generation has changed. By continuing to compare before-and-after countermeasures, it becomes easier to determine which actions led to improvements in power generation.

Especially at large power plants, a system for accurately sharing the locations of anomalies detected by remote monitoring on site is important. If you record PCS with low output, strings showing abnormalities, shaded areas, rows prone to soiling, places where water accumulates, cleaning coverage, repair locations, and inspection photos together with location information, stakeholders can more easily confirm the same spot. By combining remote monitoring data with on-site location information, it becomes easier to explain the priority of cleaning, weeding, repairs, and specialist inspections, and it also streamlines verification of recurrence in future inspections.

If you want to use remote monitoring to increase power generation and link early detection of anomalies to on-site response, utilizing LRTK is also effective. LRTK, as an iPhone-mounted GNSS high-precision positioning device, is useful for recording inspection locations within a solar power plant, equipment that has shown anomalies, areas around the PCS, places prone to soiling, locations where shadows occur, areas with poor drainage, cleaning zones, repair locations, and on-site photos together with high-precision location information. By linking anomalies found through remote monitoring to on-site records with location information, it becomes easier to pursue power generation improvements based on on-site data rather than intuition.