Six perspectives for isolating causes of low power generation using operating rate data

When you feel the power generation is low, the first thing you want to check is the power generation graph itself. However, looking only at the power generation does not allow you to determine whether the cause is insufficient solar irradiance, equipment shutdown, output curtailment, or the effects of dirt and shading. This is where it becomes important to use operating rate data to separate the time periods when the system was in a state capable of generating power from the times it was stopped or operating at low capacity. By checking the operating rate, you can more easily grasp not just fluctuations in power generation but when and to what extent the equipment was in operation during specific time periods. This article explains six perspectives for narrowing down causes from operating rate data, aimed at practitioners investigating "low power generation".

Table of Contents

• Basics of Using Availability Data to Detect Decreases in Power Generation

• Perspective 1: Check the power generation stoppage time and the low-generation time separately

• Perspective 2: Evaluate the relationship between solar radiation conditions and operating rates side by side

• Perspective 3: Reading stoppage patterns from time-of-day utilization rates

• Viewpoint 4 Narrow the scope of possible causes by differences in operating rates at the equipment-unit level

• Perspective 5: Overlay alert history and recovery times on uptime

• Perspective 6: Distinguish sustained low utilization from temporary drops

• Approach to Linking Utilization Data to On-site Verification

• Summary: Isolate the causes of low power generation in order, starting with the operating rate.

Basics of detecting declines in power generation using availability data

When investigating the causes of low power generation, it is important to first separate the result "generation was low" from the condition "whether the equipment was in a state capable of generating power." Power output fluctuates due to multiple overlapping factors such as solar irradiance, ambient temperature, shading, soiling, equipment condition, output control, and downtime. Therefore, pursuing only the power output numbers can make the range of possible causes too broad, making it difficult to prioritize on-site inspections and recovery work.

Operating rate data is an indicator used to understand how much of a given period a facility was in an operating state. The definition varies depending on the actual management dashboard or recording method, but it is generally checked based on time during which it was not stopped, time during which it was capable of generating power, time judged as operating normally via communications, and so on. It should be noted that a high operating rate does not necessarily mean that the power output is also normal. Even if the system is operating, generation can be lower than expected due to weak solar irradiance, dirty panel surfaces, shading, or output being curtailed by equipment.

Conversely, if power generation is low and the operating rate has also declined, you should prioritize checking for equipment shutdowns, communication failures, protection trips, manual stoppages, impacts from the grid, and the like. In other words, operating-rate data is not a catch-all metric that can instantly identify the cause of reduced generation, but it serves as an entry point for dividing causes into “the problem of time when generation could not occur” and “the problem where generation was occurring but output did not increase.”

In practice, it is effective to check operating rates by day, by time of day, and by individual equipment, and to cross-check them against power generation data, weather data, alert history, and inspection records. When you find a day with low power generation, don't judge based only on the day's total; check which time periods saw a drop in operating rate, whether only certain equipment showed low rates, and whether there are differences compared with the previous day or with days under similar conditions. By following this order you can more easily narrow down the scope of what needs to be checked before visiting the site.

Perspective 1: Confirm power-generation stoppage time and low-generation time separately

The first perspective when examining availability data is to separate periods of generation stoppage from periods of low generation. A state of low power output can mean either that the equipment was completely stopped, or that the equipment was running but producing low output. Because these two have different causes and require different responses, treating them both as the same "reduced generation" can easily lead to incorrect judgments.

If there is a clearly defined period of power generation stoppage, check whether there were any equipment shutdowns, grid abnormalities, protection operations, maintenance work, communication outages, or similar issues during that time. If the stoppage occurs near the daytime generation peak, even a short interruption can have a large impact on the day’s energy production. In particular, if a stoppage occurs from late morning to early afternoon on a sunny day, the total daily generation tends to decrease, and it may superficially appear as if the overall performance of the facility has declined.

On the other hand, if availability is high but only the power generation is low, consider that the system was not stopped but that there may have been conditions limiting output. Candidates include cloudy skies, rain, snow, contamination such as yellow dust or pollen, bird droppings, shadows from nearby structures, differences in output at the panel or circuit level, and output reduction due to temperature rise. In this case, simply checking stop histories will not reach the root cause. It is necessary to examine the generation curve during periods when availability is maintained and to compare with systems of the same orientation, similar capacity, and similar installation conditions.

In practice, you first check the operating rate for the target day by time slot to see if there are periods when the operating rate is zero or has fallen significantly. Next, you confirm whether power generation is low even during time slots when the operating rate is maintained. Doing it in this order allows you to separate losses due to equipment shutdowns from losses due to output reductions. The more urgently you need to isolate the cause of low generation, the more important it is to start with the major branch — whether the equipment was stopped or running — rather than immediately searching for detailed fault locations.

Perspective 2: Assess the relationship between solar radiation conditions and operating rate side by side

When you find a day with low power generation, if you look only at the operating rate you may stop at the conclusion that "the equipment was operating but the generation was low." However, in solar power generation the influence of solar irradiance is large, so unless you check the relationship between operating rate and solar irradiance side by side, you can easily confuse equipment-side problems with the effects of natural conditions.

For example, even if the operating rate is high, if it is cloudy all day and solar irradiance is low, power output will be low. In this case, it may be a natural decline due to the weather rather than an equipment fault. Conversely, if solar conditions are good but power output is low and the operating rate has also decreased, priority should be given to checking the equipment side for stoppages, protective operations, etc. If solar conditions are good and the operating rate is high but only the power output is low, suspect more detailed factors such as soiling, shading, circuit mismatch, temperature effects, or output limits.

What you should be careful about here is not to judge based solely on impressions of the weather. Even if people on site feel “it must have been sunny,” there may have been many thin clouds, solar irradiance may have been weak only in the morning, or the area immediately around the power plant may have been cloudy even if surrounding regions were sunny. If possible, it is preferable to check solar irradiance data or meteorological records near the plant and compare them with power generation on the same time axis. Even if solar irradiance data are not available, you can establish a benchmark for whether a condition is abnormal by comparing with nearby facilities of similar scale or with generation records from past days that had similar weather.

Operating rate data become meaningful when combined with solar irradiance conditions. When the operating rate is low, it is important to know whether the system was down during periods of strong irradiance. If it stopped during periods of weak irradiance, the impact on power generation is limited, but if it stopped during periods of strong irradiance, the losses are large. When isolating the causes of low generation, checking whether the periods of reduced operating rate coincide with periods of strong irradiance makes it easier to determine the priority of responses.

Perspective 3 Inferring stoppage patterns from time-of-day utilization rates

The daily operating rate alone may not be enough to sufficiently narrow down the causes of decreased power generation. For example, even if a day's operating rate is displayed as 90%, the impact can vary greatly depending on whether the 10% downtime occurred in the early morning, around noon, or in the evening. In solar power generation, because solar irradiance varies by time of day, the effect on power output is not the same even for the same downtime.

Looking at operating rates by time of day, patterns of stoppages and low operation become apparent. If startup is delayed only in the morning, possible causes include startup conditions, recovery from nighttime shutdowns, delays in communication updates being applied, and the effects of morning shadows. If the operating rate suddenly drops around midday, you should check equipment protection actions, temperature rises, output control, and grid-side fluctuations. If it stops early only in the evening, factors such as pre-sunset shadows, configured conditions, equipment shutdown detection, and changes in the surrounding environment may be involved.

Also, it's important to know whether the operating rate drops at the same time each day or only on specific days. If it occurs at the same time each day, environmental factors, configuration or settings, or the effects of routine/scheduled controls are more likely suspects. On the other hand, if it occurs sporadically, candidates include equipment failures, transient grid/system anomalies, sudden weather changes, or stoppages due to maintenance or operational work. When investigating the causes of low power generation, check not only single-day anomalies but also time-of-day patterns over several days to several weeks, as this makes it easier to separate random fluctuations from recurring problems.

Time-of-day availability is also useful for scheduling on-site inspections. For example, if a problem only causes a drop in the morning, visiting the site in the afternoon may make it hard to see the cause. Time-dependent issues such as shadows or delayed startups are difficult to diagnose unless checked during the period they occur. By identifying the times of decreased availability from the data before conducting on-site checks, you can make more effective use of limited inspection time.

Perspective 4: Narrow the range of causes by differences in equipment-unit operating rates

When isolating the causes of low power generation, it is also important to examine differences in operating rates at the equipment level. Even if the plant's overall output is low, not all equipment necessarily declines to the same extent. Determining whether the reduction is common across the whole plant, confined to certain power conditioners, or limited to specific circuits or zones can greatly narrow the scope of the cause.

If all units show similarly low operating rates, prioritize checking factors that affect the entire power plant. For example, an outage on the grid side, impacts on the power receiving equipment, overall output control, failures in the communications system, and weather conditions are possible candidates. If a drop occurs across the plant at the same time, it is more efficient to look for common causes before suspecting individual pieces of equipment one by one.

On the other hand, when only part of the equipment has a low operating rate, prioritize checking the devices, circuits, sections, communication paths, and protective devices connected to that equipment. If, for example, only a particular power conditioner has a long downtime, only circuits related to the same racking row have low values, or only equipment belonging to the same collection system shows abnormalities, it becomes easier to narrow down the scope of the cause. At this stage, proceed to check the target equipment’s alert history, on-site equipment condition, wiring connections, breaker status, fuses, terminals, and nearby shading and dirt.

When comparing on a per-equipment basis, it's important to align capacity and installation conditions. If you simply compare power generation between systems with different capacities, a system may appear to be underperforming even though the difference is normal. Comparing by operating rate reduces the influence of capacity differences, but if installation orientation, tilt, shading patterns, equipment specifications, or the number of connected circuits differ, you may not be able to judge them by the same criteria. It is therefore preferable to compare systems whose conditions are as similar as possible.

When power generation is low, looking only at the overall total obscures the range of possible causes. By checking unit-level operating rates for each piece of equipment, you can determine early whether it’s a system-wide issue or a localized problem. Being able to make that distinction makes it easier to narrow down which panels and equipment to check on site and reduces inspection rework.

Viewpoint 5 Overlay alert history and recovery times on uptime

Once a drop in uptime is confirmed, the next things to check are the alert history and the recovery time. Uptime data gives a broad indication of whether the equipment was running, but it does not fully explain why it stopped. Therefore, cross-check which alerts, alarms, protective actions, communication anomalies, and manual interventions were recorded during the period when the stoppage or low operation occurred.

If the time an alert occurred and the time the operating rate dropped coincide, it becomes easier to narrow down candidate causes. For example, if a record related to equipment protection appears, the operating rate then falls, and it subsequently recovers automatically, the stoppage may have been caused by a temporary change in conditions. If the stoppage continues without recovery, an on-site reset, inspection of components, or checks on the system side may be required. If communication anomalies are recorded, you need to distinguish whether the equipment actually stopped or whether only data acquisition failed.

The recovery time is also important information. Looking only at the time it stopped does not allow a full assessment of the impact on power generation. Whether it was restored quickly, was down for most of the daytime, or continued until the next day affects both the magnitude of the generation loss and the urgency of the response. If the recovery time is not recorded, or if alerts remain even though the availability has returned, check the consistency of the records. When the display on the management dashboard does not match on-site conditions, consider delays in communication or data updates.

When reviewing alert history, it is important not to conclude the cause based only on the displayed name. Even alerts of the same type can have different underlying causes depending on the conditions under which they occurred, equipment configuration, and site environment. Also, the first alert that appears is not necessarily the root cause; it may have been triggered in conjunction with another anomaly. Arranging the sequence—start of decreased operational rate, alert occurrence, output drop, recovery, and recurrence—in chronological order makes it easier to determine whether an alarm was a one-off or a recurring issue.

When investigating the causes of low power generation, the key is to overlay operating rate and alert history on the same time axis instead of viewing them separately. This allows you to determine to what extent generation stoppages actually contributed to the drop in output and whether any anomalies that require action remain.

Perspective 6: Distinguish persistent low utilization from temporary declines

When using operating-rate data to isolate causes, it's essential to distinguish whether the decline is persistent or merely temporary. If low power generation occurs on only one day, temporary factors such as weather, brief outages, maintenance work, or communication outages may be responsible. On the other hand, if the decline continues for several days or longer, or if low operating rates repeatedly occur at the same time of day, there may be ongoing problems with equipment or the environment.

With sustained low operating levels, generation losses accumulate the longer the causes are left unaddressed. For example, if a particular asset repeatedly shuts down and restarts every day, the drop may look small on a daily basis but can become a large difference in generated energy over a month. When low generation is buried in day-to-day variability, responses tend to be delayed. Looking at operating rates not only daily but also on a weekly and monthly basis makes it easier to detect chronic declines.

Even if the drop is temporary, caution may be necessary depending on the circumstances. If a stoppage occurred due to lightning, a power outage, a typhoon, strong winds, or heavy rain, an on-site inspection may be required even if recovery has taken place. This is because, even if power generation appears to have resumed, there may be loose connections, signs of water ingress, internal equipment abnormalities, or recorded operation histories of protective devices. Avoid judging safety or overall integrity solely on the fact that the operating rate has returned; instead, verify against inspection records according to the circumstances of the incident.

When judging sustained low operation, it is important to make the comparison criteria clear. Rather than relying solely on day‑over‑day comparisons, comparing with the same month in the previous year, clear‑sky days in the same season, installations of similar capacity, and installations with the same orientation makes it easier to determine whether an anomaly is present. However, if there have been equipment additions, configuration changes, changes in the surrounding environment, tree clearing or building construction, or equipment upgrades, simple historical comparisons may not be valid. Operating-rate data are easy to handle numerically, but if they are not interpreted to reflect changes in on‑site conditions, they can lead to incorrect judgments.

In practical work to investigate the causes of low power output, it is important to assess continuity rather than be swayed by isolated anomalies. By confirming whether a drop in operating rate is only temporary or whether the same equipment shows a continued decline, you can more easily distinguish cases that require urgent intervention from those that should be monitored or checked during scheduled inspections.

Approach to Connecting Utilization Data to On-site Verification

Operating-rate data is not just for guessing causes on a screen. In practice, it is used to narrow the scope of checks based on operating rates and decide what to look for on site. If you go to the field with only the vague impression that power generation is low, there will be too many places to check and it will take time. However, if you organize in advance the time periods of low operating rates, the affected equipment, the frequency of occurrences, and the alert history, it becomes easier to set priorities for inspections.

When proceeding to on-site verification, first identify the equipment whose operating rate has declined. Next, inspect the switchgear, circuit breakers, fuses, junction boxes, wiring routes, communication devices, and the surrounding environment related to that equipment. If the operating rate is low overall, the process shifts from individual units to checking common/shared equipment, the system/grid side, incoming power equipment, and overall communications. Looking only at power generation output makes this priority unclear, but using operating rate allows you to narrow the scope of possible causes step by step.

Also, it is important to always cross-check what was confirmed on site against the operating-rate data. Even if an anomaly is found in the field, you may overlook another cause if you do not verify that it corresponds to an actual reduction in power output. For example, even if soiling is found, if it is not related to the period of reduced operating rate, the primary cause may be an equipment shutdown. Conversely, if an equipment shutdown is found but the reduction in power output continues outside the shutdown period, you need to consider combined factors such as shading or soiling.

To make availability data useful for on-site inspections, the way you record information is also important. Recording the inspection date, inspection time, the equipment involved, on-site conditions, whether recovery actions were taken, and whether recurrence occurred will make it easier to judge when a similar decline happens next time. In particular, when manual intervention or temporary operations bring improvement, even if power generation has returned at that point, the root cause may still remain. Do not consider recovery of availability alone as completion; manage through including recurrence checks over a certain period to reduce missed drops in power generation.

The cause of low power generation cannot always be determined from on-screen figures alone or from visual inspection at the site alone. By using operating-rate data as the entry point and connecting power generation, solar irradiance, alerts, differences between equipment units, and on-site verification, it becomes easier to get closer to the cause.

Summary: Isolate the causes of low power generation in order, starting with the operating rate

When power generation is low, it is important not to assume the cause based solely on the total generation figure. Power output from solar photovoltaic systems is influenced by many factors, including irradiance conditions, equipment shutdowns, output degradation, soiling, shading, control systems, and communications status. Among these, availability data is an effective clue for distinguishing whether the equipment was in a condition capable of generating power or whether stoppages or low operation occurred.

First, check separately the power generation stop time and the low-generation time. Next, align solar irradiance conditions with operating rates to determine whether the drop is a natural decrease due to weather or an equipment-side issue. Further, read shutdown patterns from time-of-day operating rates and narrow the scope of possible causes by differences between equipment units. Overlaying alert histories and recovery times makes it easier to confirm the background and recurrence of the shutdowns. Finally, determine whether the decline is temporary or persistent, and link that to on-site inspection and maintenance actions.

By leveraging availability data, you can turn the cause of reduced power output from a vague “it’s somewhat low” into concrete details about which time periods, which assets, and which conditions were occurring. If you can narrow down the cause quickly, you can reduce unnecessary site checks and more easily prioritize necessary actions. To continuously manage declines in power generation, it’s essential to monitor daily generation together with availability, shutdown history, alerts, and on-site records.

To accurately understand a low power generation condition, it is important to establish a system that takes availability/operational-rate data as the starting point and continuously monitors equipment condition and generation performance. By accumulating daily records and enabling generation output, availability (operational rate), solar irradiance conditions, alert history, and on-site inspection results to be reviewed on the same timeline, it becomes easier to isolate causes and connect them to subsequent countermeasures.