Systems and methods for energy production monitoring current transformer issue detection

Abstract

A system and method for detecting current transformer (CT) installation issues in energy assets are described. The system retrieves production and consumption data from energy assets, calculates a consumption correlation value and a charge ratio, and compares these metrics to predefined thresholds. A CT issue is identified when the correlation value exceeds a similarity threshold or the charge ratio falls below a minimum threshold.

Description

SYSTEMS AND METHODS FOR ENERGY PRODUCTION MONITORING CURRENT TRANSFORMER ISSUE DETECTIONCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent App. No.63 / 750,222, filed January 27, 2025, which is hereby incorporated by reference herein. TECHNICAL FIELD

[0002] This invention relates generally to current transformers, and more particularly to issue detection for current transformers.BACKGROUND

[0003] Current transformers ("CTs") are widely used to measure electrical current in distributed energy systems, including residential and commercial photovoltaic (PV) installations and hybrid solar-plus-storage systems. In these systems, CTs provide consumption and net-metering signals that are essential for system monitoring, performance verification, and the control logic of attached energy-storage devices. For example, many battery inverters rely on CT measurements to determine when to charge or discharge energy in response to site load behavior.

[0004] However, CTs are frequently installed incorrectly in the field due to factors such as limited inspection time, inconsistent installer training, complex wiring layouts, and the lack of immediate feedback on CT orientation or conductor placement. Incorrectly installed CTs can produce inverted signals, artificially inflated or deflated consumption readings, or flat-line data that lead to a number of operational issues.BRIEF DESCRIPTION OF DRAWINGS

[0005] Various needs are at least partially met through provisions of the systems and methods for current transformer issue detection described in the following detailed description, particularly when studied in conjunction with the drawings. A full and enabling disclosure of the aspects of the present description, including the best mode Attorney Docket No. 21830-160618-USthereof, directed to one of ordinary skill in the art, is set forth in the specification, which refers to the appended figures, in which:

[0006] FIG. 1 is a block diagram of a detection system in accordance with some embodiments;

[0007] FIG. 2 is a flow diagram of a CT detection system in accordance with some embodiments; and

[0008] FIG. 3 is a block diagram of an exemplary computing system in accordance with some embodiments.

[0009] Elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions and / or relative positioning of some of the elements in the figures may be exaggerated relative to other elements to help to improve understanding of various embodiments of the present teachings. Also, common but well-understood elements that are useful or necessary in a commercially feasible embodiment are often not depicted in order to facilitate a less obstructed view of these various embodiments of the present teachings. Certain actions and / or steps may be described or depicted in a particular order of occurrence while those skilled in the art will understand that such specificity with respect to sequence is not actually required.DETAILED DESCRIPTION

[0010] Generally speaking, various embodiments described herein provide systems and methods for detecting issues in a current transformer (CT) that is coupled to an energy production device such as a photovoltaic (solar) system or an energy storage system. The issue detection systems and methods retrieve production data (e.g., power generated) and consumption data (e.g., power consumed or drawn by loads / storage) from an energy asset, then run statistical analyses to determine if the CT is incorrectly installed or malfunctioning. By automating these analytics, the system can rapidly identify CT installation errors that would otherwise require time-consuming manual data inspection, thereby improving detection accuracy, consistency, and operationalAttorney Docket No. 21830-160618-USefficiency in managing solar assets. For example, solar energy production sites (especially those with battery storage) frequently suffer CT-related issues (estimated in 1 out of 10 to 1 out of 3 sites), and prior approaches to CT issue detection have been predominantly manual and prone to human error. The present automated CT issue detection system enabling accurate, low-latency issue detection without manual intervention, providing a practical application that addresses a hardware-rooted problem.

[0011] In some embodiments, a current transformer issue detection system (e.g., system 100 in FIG. 1) is provided. The system includes at least one processor and a non- transitory machine-readable medium storing instructions. When executed by the processor, these instructions cause the system to detect issues with an installed CT. At a high level, the processor is configured to receive production data and consumption data associated with an energy asset that has a monitored current transformer. Using these inputs, the processor calculates a consumption correlation value that represents the degree of similarity or correlation between the production data and the consumption data. In general, this consumption correlation value can be computed as a statistical measure (for example, a Pearson correlation coefficient) reflecting how closely changes in the production signal correspond to changes in the consumption signal. The processor also calculates a charge ratio based on the production and consumption data. The charge ratio represents the percentage or fraction of the energy produced by the asset that is available to charge an associated storage device (such as a battery). In essence, the charge ratio quantifies how much of the generated power is not being immediately consumed on-site and thus could be used to charge storage; a correctly installed CT should report consumption data such that this ratio stays within a normal range.

[0012] After computing these metrics, the processor compares the consumption correlation value to a predetermined correlation threshold and compares the charge ratio to a predetermined charge ratio threshold. These thresholds define the bounds of normal behavior: for instance, the correlation threshold might be a high value (e.g.,Attorney Docket No. 21830-160618-US0.97) such that exceeding it indicates an unusually strong correlation between production and consumption, and the charge ratio threshold might be a minimum acceptable value (e.g., 5%) below which too little of the production is accounted for as charging capacity. Based on the comparisons, the system determines that a CT issue has occurred if one or more of the metrics indicate abnormal conditions. In particular, an issue is flagged when the consumption correlation value exceeds its correlation threshold or when the charge ratio falls below its minimum charge ratio threshold. Either of these conditions suggests a possible mis-installation or malfunction of the CT- for example, a very high correlation between production and consumption data may indicate that the CT is wired incorrectly (e.g., measuring the wrong quantity or reversed), and a very low charge ratio means the CT-reported consumption is so high that almost no production is left for charging, which could also indicate incorrect CT readings. In summary, the system's logic tests whether the production and consumption data relationships deviate from expected norms, and if so, it automatically identifies a CT issue and generates an alert or report for that condition.

[0013] To support different deployment scenarios, the CT issue detection system can be implemented in various forms. For example, in some embodiments, the above functionality is implemented as a software application running on a server or cloud platform that monitors solar installations; in other embodiments, it may be built into a local controller at the site. The instructions executed by the processor can reside on any suitable non-transitory machine-readable medium (such as a flash memory, hard drive, or solid-state disk), ensuring the system's logic is persistently stored and not merely a transitory signal. It should also be appreciated that the invention encompasses corresponding methods of CT issue detection and computer-readable media with stored instructions, consistent with the system features described.

[0014] In this description, the technical terms are used consistently with their ordinary meanings in the field, except where explicitly defined. For clarity, as used herein, the word "or" is intended to be inclusive, meaning "and / or" unless the context clearlyAttorney Docket No. 21830-160618-USdictates otherwise. Thus, a condition of "A or B" is satisfied by A alone, B alone, or both A and B. The singular forms "a", "an", and "the" include plural references unless the context clearly dictates otherwise. Additionally, when we refer to components being "coupled" or "attached," this encompasses both direct and indirect connections (through intermediate elements) unless otherwise specified. Approximating language such as "about" or "approximately" when used to modify a quantitative value means that minor variations (e.g., within 10%) of the stated value are within the intended scope.

[0015] Referring now to FIG. 1, a block diagram of an example CT issue detection system 100 is shown. The system 100 is designed to monitor one or more energy assets 101, such as photovoltaic (PV) solar panel systems, energy storage units (batteries), or combined solar-plus-storage installations. Each energy asset 101 typically has one or more current transformers (CTs) connected to its electrical lines, which measure parameters like current flowing to loads or storage. The CTs enable remote monitoring of the asset's power production and power consumption. Production data generally represents the power output of the solar panels (or other generation source), while consumption data represents the power usage or the portion of production being consumed on-site (including by any battery charging). When a CT is installed correctly, these two data streams (production vs. consumption) should reflect a normal expected relationship; if the CT is mis-installed (for example, clamped in the wrong direction or on an incorrect conductor), the consumption readings become unreliable. As described earlier, such CT errors can cause operational issues, including improper battery charge / discharge behavior.

[0016] In some embodiments, the CT issue detection system is configured to identify specific categories of CT installation and configuration errors that can occur in residential and commercial solar-plus-storage installations. These potential error modes include, but are not limited to: a CT installed on the incorrect electrical phase; a CT installed on photovoltaic (PV) production conductors instead of the conductorsAttorney Docket No. 21830-160618-USsupplying home or site consumption; a CT installed in a reversed orientation resulting in inverted polarity measurements; a CT that is not fully closed around the conductor or is missing entirely from the wiring; the presence of additional PV systems or generation sources that are not included within the CT measurement path; and incorrect CT or meter configuration parameters such as an improperly set CT ratio, which may cause measured current values to appear artificially high or low. The detection algorithms described herein, including correlation analysis, charge-ratio evaluation, consumption plausibility checks, and missing-data detection, are capable of detecting one or more of these error states based on the statistical and physical inconsistencies they produce in the reported production and consumption data.

[0017] In system 100, the energy asset 101 generates raw data which is collected and processed through several components. A data aggregator 102 (which may be a third- party monitoring platform or data service) gathers the energy data from the asset. This data typically includes time-series readings of production and consumption. The system 100 includes an API ingestion component 114 that retrieves the energy asset data from the data aggregator(s) 102 via one or more Application Programming Interfaces (APIs). In some embodiments, multiple API ingestion components 114 are provided to interface with different external data sources (for example, different inverter manufacturers' portals or metering systems). The data fetched through API may consist of raw production time-series data 118A and raw consumption time-series data 118B. These raw datasets might be in the form of detailed timestamped logs or tables, possibly containing noise or irregularities. The system can process or refine the raw data into transformed production time-series data 120A and transformed consumption timeseries data 120B. This refining step may involve cleaning the data, aligning timestamps, filtering out anomalies, and generally preparing the data for analysis. In practical implementations, this data processing can be structured in pipelines; for example, the raw data (sometimes called "bronze" data) may be stored in one stage, and then transformed into "silver" tables for analysis. The transformed production andAttorney Docket No. 21830-160618-USconsumption data 120A, 120B are stored in a data storage 104, which can be any suitable database or data repository (local or cloud-based) capable of handling timeseries data. Certain data management policies may be applied in storage 104, for instance, raw consumption data might be periodically refreshed and not stored beyond a set retention period (e.g., 30 days) to conserve space. The data storage 104 may also maintain an alert history tracking component 117 that logs past CT issues detected by the system. This historical log can be used for later review, auditing, or to inform maintenance strategies.

[0018] In some embodiments, the transformation from the raw time-series data 118A / 118B to the transformed datasets 120A / 120B is purpose-built for CT diagnostics. The pipeline performs (i) normalization across heterogeneous inverter / aggregator formats, (ii) timestamp reconciliation to a common cadence, (iii) removal of telemetry gaps and spurious spikes caused by communications jitter, and (iv) derivation of CT-specific features (e.g., signed surplus energy and bounded standby consumption). These operations materially improve the functioning of the computing system by reducing noise and ensuring that the CT issue detector 106 operates on stable, sensor-appropriate inputs, rather than on generic or unstructured data.

[0019] The CT issue detector 106 is a software and / or hardware module (executing on one or more processors, as illustrated in FIG. 3) that implements the CT issue detection algorithm. It retrieves the necessary time-series data from the data storage 104 - specifically, the transformed production data 120A and transformed consumption data 120B - and compiles them into a CT detection-specific dataset or table for analysis. This preparation can include aggregating data over a relevant time window, aligning production and consumption readings, de-duplicating records, and ensuring the dataset is complete and clean for the detection algorithm. Once the dataset is prepared, the CT issue detector 106 executes the detection algorithm to determine whether the current transformer associated with that energy asset 101 is experiencing an issue. The detection algorithm's logic mirrors the steps outlined earlier (and described in detailAttorney Docket No. 21830-160618-USwith reference to FIG. 2): it performs a correlation analysis between the production and consumption signals, evaluates a charge-ratio metric, checks for missing or insufficient consumption data, and looks for implausible consumption values that might indicate a problem. By analyzing these aspects, the CT issue detector 106 can identify if the CT is likely mis-installed or malfunctioning.

[0020] If the CT issue detector 106 finds that an issue condition is met (for example, the correlation is above the threshold or the charge ratio is abnormally low, or other criteria described below), it generates an outcome of CT detection which typically includes the classification of the issue (type of fault) and relevant diagnostics. This outcome is sent to an alert message queue 112. The alert message queue 112 serves as a buffer and routing mechanism for issue alerts: it stores alerts as discrete messages or records, possibly tagging them with timestamps, site identifiers, issue types, and priority levels. The use of a queue allows the detection system to operate asynchronously with downstream handling - the detection can continue analyzing new data while alerts are processed in parallel.

[0021] The system 100 further includes an alert handling system 124, which retrieves and processes alerts from the message queue 112. The alert handling system 124 may comprise a notification engine and a rules-based triage component. Its role is to route CT issue alerts to the appropriate personnel or systems and to initiate remediation workflows. For instance, the alert handling system 124 can be configured to send out notifications to users (such as solar operations engineers or maintenance teams) when a CT issue is detected. Notifications can be delivered in various forms (for example, an automated email or SMS, a push notification in a monitoring app, or an entry on a dashboard) and may include details of the issue and suggested actions. Additionally, the alert handling system 124 applies filtering or routing logic: for critical issues or specific types of faults, it may escalate alerts to certain recipients (e.g., a senior technician or installer responsible forthat site), whereas less urgent alerts might be logged for later review.Attorney Docket No. 21830-160618-US

[0022] Integrated with alert handling is a field technician remediation scheduling service 126. This component uses the alert information to schedule maintenance or corrective actions for the affected energy asset 101. In one embodiment, when the alert handling system 124 determines that a CT issue requires on-site correction, it forwards the alert (or a summarized report of the issue) to the scheduling service 126. The scheduling service 126 then creates a remediation task for a field technician, potentially prioritizing and grouping tasks if multiple issues are detected. For example, if the CT issue detector flags several sites with CT problems, the scheduling service might arrange those tasks in an optimized route for technicians, prioritize them based on severity (with completely non-reporting CTs fixed before minor calibration issues), and include relevant details in each task description (site location, issue type, recommended fix). In some embodiments, a detected issue may automatically trigger a control signal to the energy asset to offline the energy asset and / or prevent the energy asset from reporting data to other assets onsite that may respond to faulty data.

[0023] In some embodiments, the combination of the alert queue 112, alert handling system 124, and scheduling service 126 may allow the system to not only detect CT issues but also to automate the response, ensuring that identified issues are promptly communicated and addressed in the field. This end-to-end flow, from detection to actionable maintenance scheduling, implements the issue reporting automation and improved efficiency in addressing CT issues as noted above.

[0024] Referring next to FIG. 2, a flow diagram of an example process 200 for CT issue detection is shown. This process 200 illustrates a method that can be performed by the system 100 (for example, by the CT issue detector 106 in conjunction with data ingestion components) to analyze the data and detect CT anomalies. The steps of process 200 can be executed by one or more processors (such as the processor in FIG. 3) running the instructions stored in memory. It should be understood that while FIG. 2 shows these steps in a particular order, some steps may be performed in parallel or in aAttorney Docket No. 21830-160618-USdifferent sequence, and some implementations might omit certain steps or add additional ones, as long as the core logic is implemented.

[0025] In step 201, the system aggregates energy asset data. This corresponds to collecting the production and consumption measurements for the asset over a relevant time period. As noted earlier, data aggregation may be accomplished via the data ingestion component 114, which interfaces with external data sources (through APIs) to pull the latest readings. In practice, the system will fetch raw data, then separate and refine it into distinct datasets: a production dataset 215 and a consumption dataset 210 for the energy asset. At this stage, the data from the asset's monitoring equipment is prepared for analysis; for example, the system may retrieve data points representing the solar production power (in watts) and the site's net consumption power (in watts), each timestamped, covering the same period (such as the last 24 hours). This retrieval may be done programmatically through a service or database call, such as via an API.

[0026] In step 221, the system evaluates the consumption data 210 to identify data points that show unexpectedly low or anomalous consumption values. The system applies a predefined consumption threshold to the consumption values. This threshold is chosen based on expected operating conditions of the asset. For example, the threshold may be set at 0 watts (zero) or a similarly low value, because a wellfunctioning residential solar system typically reports at least some nominal consumption (even if minimal, like standby power draw). In many solar-plus-storage installations, a consumption reading of exactly 0 W or a negative value is not physically meaningful under normal conditions (since even when home loads are zero, the battery or inverter typically draws a small amount of power for operation). Thus, the system flags any consumption data points at or below 0 W (or a set threshold) as indicating an abnormal condition. This check addresses situations such as a CT installed backwards (which can result in negative readings) or a CT not properly sensing current (yielding zero when there should be load). As an example, many solar systems maintain a standby load of at least ~500 W, so a reported 0 W consumption is indicative of an issue.Attorney Docket No. 21830-160618-US

[0027] In step 222, the system determines how frequently the above anomaly occurs by counting the number of consumption data points that are at or below the consumptionvalue threshold, and comparing that count to a predetermined count threshold. The count threshold could be a number like ten occurrences within a day (or within a set of consecutive readings). This step differentiates a single sporadic glitch from a persistent issue. For instance, one or two zero readings might be due to a temporary communication drop or a brief sensor hiccup, whereas a larger number (e.g., 10 or more zero readings in 24 hours) strongly suggests a real CT installation problem. If the number of unexpected consumption readings exceeds the predetermined count threshold, the system interprets this as indicative of a CT issue. In other words, when there are too many low / zero consumption data points, the processor will determine that a CT issue has occurred and proceed to step 227. By performing steps 221 and 222, the system can catch scenarios where the CT readings are frequently zero or nonsensical, which is a strong symptom of a mis-wired or non-reporting CT.

[0028] In some embodiments, the system may further verify whether consumption data is present at all for the site. For example, the system may check if any consumption data 210 has been received from the data ingestion component within an expected time interval. If the energy asset 101 includes a storage system or is otherwise supposed to provide consumption readings, then a complete absence of consumption data in a given period (for example, no data received for several hours or days) may indicate an issue.

[0029] In step 223, the system calculates the consumption correlation value between the consumption data 210 and the production data 215. In one embodiment, the consumption correlation value is computed as a Pearson correlation coefficient over a set of paired production and consumption data points. This coefficient (ranging from -1 to +1) measures linear correlation: a value near +1 means the two datasets rise and fall together very closely. The system expects that in a normal, correctly installed CT scenario, the site's net consumption does not perfectly track production - for example, when solar production spikes at midday, consumption might not spike identicallyAttorney Docket No. 21830-160618-USbecause some energy goes into charging the battery or serving loads, making the consumption pattern different from the production pattern. However, if a CT is installed incorrectly (say, on the solar output instead of the feed to the loads, or wired such that production is being counted as "consumption"), then the reported consumption might mirror the production curve, yielding a high correlation. Thus, a high correlation value is a telltale sign of a CT issue. The system's calculated consumption correlation value is then used in step 224, where it is compared to a predetermined correlation threshold. For example, the threshold might be set at 0.97. If the correlation value meets or exceeds this threshold, the system identifies a CT issue and proceeds to step 227.

[0030] In step 225, the system computes the charge ratio utilizing the production data 215 and consumption data 210. The charge ratio is defined to capture the fraction of generated energy that is not immediately consumed and thus is available for charging an associated energy-storage device, such as a battery. One way to calculate the charge ratio is to compute, for each time interval, the difference between production and consumption (which represents surplus energy that could charge the battery) and then divide that by the production to get a percentage. Aggregating this over a time window yields an average charge availability. For example, over a day, the system might integrate total production vs. total consumption and see what percentage of production was left unused by loads (hence potentially used to charge storage). The resulting charge ratio provides an indication of whether the CT is accurately capturing consumption behavior while low or inconsistent charge-ratio values may suggest that the CT is misconfigured or improperly installed.

[0031] In step 226, the system compares the charge ratio to a predetermined charge ratio threshold. The threshold might be set as a minimum acceptable charge ratio, for instance 5% (0.05) in one embodiment. When the charge ratio falls below the minimum threshold, a CT issue may be identified and an alert and / or report generated in step 227 described below.Attorney Docket No. 21830-160618-US

[0032] In some embodiments, the correlation analysis (steps 223-224), the charge ratio analysis (steps 225-226), and the consumption data checks (steps 221-222) may run in parallel with as indicated in FIG. 2. The system can detect an issue via any one of these pathways or a combination. For example, one CT mis-wiring scenario might produce both a high correlation and multiple zero-reading occurrences, triggering two conditions, whereas another scenario might only trigger the charge ratio condition. The system is designed to capture any one or more of these symptoms. In practice, as soon as one of the issue criteria is satisfied (a threshold condition is met), the process can flag an issue without necessarily waiting for the others.

[0033] In step 227, upon detecting a CT issue via one or more of the foregoing steps, the system generates an alert for the issue. This step consolidates the findings into a human or machine-readable format that can be communicated out. For example, the report may specify the type of issue identified (e.g., installation error, configuration error, or data-quality anomaly), the relevant diagnostic metrics that triggered the detection (such as correlation threshold violations, abnormal charge-ratio values, or missing consumption data), and / or a prioritized list of recommended remedial actions to be performed on the affected current transformer.

[0034] In some embodiments, the report or alert is automatically transmitted to the appropriate parties. For example, the system may automatically email the report to the operations team responsible for that asset, or send an alert to a maintenance dashboard where it appears alongside other site alarms. The report could also be accessible via a user interface that allows an operator to review the CT issues across all assets. Additionally, as part of step 227 or immediately after, the system can initiate a maintenance scheduling operation, such as integrating with the scheduling service 126 to create a work order for the issue. The maintenance coordination and scheduling service may then assign and schedule field tasks based on factors like issue type, location, severity, etc. This automated initiation of maintenance ensures that once anAttorney Docket No. 21830-160618-USissue is detected, it moves swiftly into the queue of tasks for technicians, without waiting for manual review.

[0035] In some implementations, the CT issue detection system can be used not only during the commissioning test of a new solar asset installation but also for real-time monitoring of the deployed asset. For example, right after a solar-plus-storage system is installed, the system 100 can run process 200 on the initial data to confirm that the CT is correctly installed before the system is handed off as operational (catching any installation mistakes early). Furthermore, the system can aggregate statistics about CT installations, such as tracking the frequency of CT issues by installer or by equipment type, and generate reports that help identify patterns (e.g., if a particular installer has a higher rate of errors, or a certain CT model tends to fail). This feedback can be used to improve training or hardware choices, thereby reducing CT issues in the future.

[0036] In some embodiments, the CT issue detection process may be an ongoing monitoring routine. The steps illustrated in FIG. 2 (or a subset of them) can be performed periodically or continuously after the CT's installation. For instance, the system might run the analysis daily, weekly, or in real-time streams, updating the correlation and charge ratio calculations with each new batch of data. In one embodiment, the instructions cause the processor to periodically retrieve the latest production and consumption data and recalculate the consumption correlation value and charge ratio for each period. By doing so, the system can detect any CT issues that develop over time such as due to equipment drift, wiring coming loose, firmware updates causing reporting issues, etc. This ongoing monitoring capability ensures longterm reliability of CT data and timely detection of issues throughout the asset's operational life. Each time the periodic check runs, steps analogous to 221-227 may be repeated, and if an issue is found, a new alert / report will be generated and maintenance scheduled as described.

[0037] In some embodiments, the CT issue detection system is further configured to automatically detect when a previously identified CT issue has been resolved. AfterAttorney Docket No. 21830-160618-USgenerating an alert corresponding to a detected issue, the system continues to execute the CT evaluation logic on a periodic or continuous basis. As additional production and consumption data become available, the system evaluates a sliding window of recent data points to determine whether any of the CT issue criteria remain satisfied. If the system determines that none of the issue-detection conditions are present within the evaluation window, the processor concludes that the CT has returned to a normal operating state. In response, the system automatically closes the previously generated CT issue alert, updates any associated alert or ticketing systems, and records a resolution event in the alert history tracking component. This automatic closure functionality enables the system to verify the success of technician-performed remediation in the field and ensures that stale or resolved CT issues are not retained in the active alert queue.

[0038] In some embodiments, the CT issue detection system operates continuously or at scheduled intervals and evaluates CT health using a rolling or sliding time window of production and consumption data. At each execution cycle, the system publishes an event indicative of the CT's current state. If one or more CT issue conditions are detected within the sliding window, the system publishes a "CT issue present" event that may trigger initial alert creation or update the status of an existing alert.Conversely, if none of the CT issue conditions are detected within the evaluation window, the system publishes a "CT issue absent" event. The "issue absent" event may be used to confirm proper functioning of the CT after installation, after corrective action has been performed, or during ongoing monitoring of operational assets. By producing both positive and negative state events, the system provides explicit visibility into transitions between normal and abnormal CT behavior.

[0039] In some embodiments, when a CT issue has previously been identified and associated with an active alert or maintenance ticket, the automatic "issue absent" detection acts as a confirmation mechanism following field technician intervention. After a technician corrects wiring, replaces hardware, or otherwise resolves theAttorney Docket No. 21830-160618-USunderlying cause, the system accumulates a sufficient amount of post-fix data within the sliding window. Once the system determines that the CT satisfies all expected behavior criteria, such as reasonable charge ratio values, acceptable consumption correlation levels, plausible consumption readings, and presence of required data, the system automatically verifies remediation and transitions the issue to a closed state. This eliminates the need for manual review of telemetry data to confirm repair effectiveness and ensures timely and accurate closure of CT-related maintenance tasks.

[0040] In some embodiments, the CT issue detection system is employed as part of an automated onboarding workflow for newly added energy assets. When a new site is integrated into the monitoring platform, the system executes a suite of automated diagnostic checks that includes CT issue detection along with other photovoltaic and performance-related validation tests. If the new site satisfies all CT-related criteria and no CT issue conditions are detected during analysis of initial telemetry, the system automatically designates the site as successfully onboarded without requiring human intervention. This approach ensures that CT mis-installation problems are identified and corrected before the system enters normal operation, improving asset reliability and reducing post-onboarding maintenance events.

[0041] In some embodiments, certain newly onboarded assets may rely on external data sources or inverter platforms that do not provide the necessary data fields for automated CT detection to execute. When the processor determines that required telemetry is unavailable for one or more CT checks, the system flags the site as requiring manual review. In such scenarios, a human operator or installer may verify the CT installation through direct inspection or by analyzing alternative data sources. Once verification is completed and the CT is confirmed to be operating correctly, the system updates the onboarding workflow accordingly. This fallback mechanism ensures reliable onboarding even when automated diagnostics cannot be performed.

[0042] Finally, referring to FIG. 3, an exemplary hardware configuration of a computing system 300 suitable for implementing the CT issue detection system is illustrated. ThisAttorney Docket No. 21830-160618-UScomputing system 300 can represent the server or device on which the CT issue detector 106 and related components execute. In the depicted embodiment, the system 300 includes a processor 302, a memory 304 (which may store the instructions 306 that implement the algorithms described above), and a storage 308 for persistent data storage. These components communicate over a system bus 310. The computing system 300 may also include an interface 312 and various input / output ports 314 for communication with external systems or user interfaces. For example, the interface 312 could include network interface hardware for connecting to the internet or cloud services (allowing the API ingestion to function, for instance). A transceiver 316 may also be present for wireless communications. The processor 302 can be any suitable processing unit (CPU, microcontroller, etc.) capable of executing software instructions; it carries out the steps of the CT detection method by executing the stored program in memory. The memory 304 and storage 308, which may include both volatile and nonvolatile memory, store the code and data needed for the CT issue detection; for example, memory 304 may hold the running program and recent data, while storage 308 might hold historical data, configuration settings (like the thresholds), and so on. When the instructions 306 are executed by processor 302, the computing system 300 performs the various operations described herein (such as data retrieval, calculation of correlation and charge ratio, comparisons, generating alerts, etc.). It should be understood that this hardware description is one example; the system could be distributed (with different components on different machines) or cloud-based, but in all cases there is at least one processor and memory executing the described functions.

[0043] In some embodiments, the system may be implemented remote of the monitored CT using only consumption and production data available with remote monitoring. In some embodiments, portions of the CT issue detection systems and methods may execute proximate to the CT hardware on a site gateway, inverter controller, or metering interface. An embedded implementation can streamtime-aligned production and consumption readings and perform the correlation,Attorney Docket No. 21830-160618-UScharge-ratio, and plausibility checks in situ to suppress mis-charging or mis-discharging behavior until the CT is corrected.

[0044] The disclosed embodiments improve the accuracy and reliability of CT-derived consumption measurements, and the efficiency and scalability ofcomputer-implemented monitoring platforms by transforming heterogeneous telemetry into CT-specific issue detection. The result is reduced false positives and false negatives relative to manual inspection and heuristic checks, faster remediation of field wiring errors, and improved long-term asset performance.

[0045] Further aspects of the disclosure are provided by the subject matter of the following clauses:

[0046] A current transformer (CT) issue detection system comprising a processor and a machine-readable medium storing instructions that, when executed, cause the processor to receive production data and consumption data associated with an energy asset coupled to a CT; calculate a consumption correlation value representing a similarity between the production data and the consumption data; calculate a charge ratio representing a percentage of produced energy available for charging a storage device; compare the consumption correlation value to a predetermined correlation threshold; compare the charge ratio to a predetermined charge ratio threshold; and determine that a CT-related issue has occurred when the consumption correlation value exceeds the predetermined correlation threshold or when the charge ratio is below the predetermined charge ratio threshold.

[0047] The system of any preceding clause, wherein the instructions further cause the processor to translate the consumption correlation value into a binary score indicative of whether the correlation meets the predetermined correlation threshold.

[0048] The system of any preceding clause, wherein the charge ratio is computed by determining differences between production data and consumption data over a defined time window and calculating an average value indicative of energy available for charging.Attorney Docket No. 21830-160618-US

[0049] The system of any preceding clause, wherein the processor determines whether consumption data is present prior to calculating the consumption correlation value and the charge ratio, and, in response to an absence of consumption data, determines that a CT issue has occurred.

[0050] The system of any preceding clause, wherein the processor compares a number of data points in the consumption data to a predetermined consumption threshold and determines that the CT issue has occurred when the number of data points at or below the threshold exceeds a predetermined count threshold.

[0051] The system of any preceding clause, wherein the processor retrieves asset data via an application programming interface and refines the asset data into the production data and consumption data used for CT issue detection.

[0052] The system of any preceding clause, wherein the processor periodically monitors the production data and consumption data after installation of the CT and recalculates the consumption correlation value and the charge ratio.

[0053] The system of any preceding clause, wherein the processor generates a report including details of the CT issue, the report comprising a prioritized list of tasks to be performed based on the type of issue detected.

[0054] The system of any preceding clause, wherein the CT issue corresponds to an installation or configuration error resulting in inaccurate consumption readings for an energy asset having an energy storage device, the inaccurate readings being associated with improper charging or discharging of the storage device.

[0055] The system of any preceding clause, wherein the processor initiates a maintenance scheduling operation in response to determining that a CT issue has occurred.

[0056] A non-transitory machine-readable medium storing instructions that, when executed, cause a processor to receive production and consumption data associated with an energy asset; calculate a consumption correlation value and a charge ratio; compare each to respective thresholds; and determine the occurrence of a CT issueAttorney Docket No. 21830-160618-USwhen the correlation value exceeds the correlation threshold or the charge ratio is below the charge ratio threshold.

[0057] The medium of any preceding clause, wherein the processor translates the consumption correlation value into a binary score.

[0058] The medium of any preceding clause, wherein the processor determines whether consumption data is present prior to calculating the metrics and determines a CT issue when the consumption data is absent.

[0059] The medium of any preceding clause, wherein the processor compares a number of data points in the consumption data at or below a predetermined consumption threshold to a predetermined count threshold and determines that a CT issue has occurred when the count exceeds the threshold.

[0060] The medium of any preceding clause, wherein the processor generates a report including a prioritized list of tasks associated with the detected CT issue.

[0061] A method comprising receiving production and consumption data associated with an energy asset; calculating a consumption correlation value; calculating a charge ratio; comparing each metric to its respective threshold; and determining that a CT issue has occurred when the number of consumption data points at or below a predetermined consumption threshold exceeds a predetermined count threshold.

[0062] The method of any preceding clause, further comprising translating the consumption correlation value into a binary score indicative of whether the correlation meets the predetermined threshold.

[0063] The method of any preceding clause, further comprising determining whether consumption data is present prior to calculating the metrics and determining a CT issue when the consumption data is absent.

[0064] The method of any preceding clause, further comprising comparing a number of consumption data points at or below a predefined consumption threshold to a predetermined count threshold and determining that a CT issue has occurred when the count exceeds the threshold.Attorney Docket No. 21830-160618-US

[0065] The method of any preceding clause, further comprising generating a report including details of the CT issue and a prioritized list of tasks to be performed based on the type of issue detected.

[0066] Those skilled in the art will appreciate that the embodiments described above can be modified or varied without departing from the scope of the invention. The specific thresholds (correlation values, charge ratio percentages, count of zero readings, etc.) can be tuned for different types of installations. Additional statistical tests or machine learning models could be incorporated to enhance detection. The architecture can scale to fleets of thousands of assets, with parallel processing of data. All such modifications are considered within the ambit of these teachings, as the invention is defined by the following claims and their equivalents, rather than the specific examples given in this description.Attorney Docket No. 21830-160618-US

Claims

What is claimed is:

1. A current transformer issue detection system comprising:a processor; anda machine readable medium storing instructions that, when executed by the processor, cause the processor to:receive production data and consumption data associated with an energy asset coupled to a current transformer;calculate a consumption correlation value between the production data and the consumption data, the consumption correlation value representing a similarity between the production data and the consumption data;calculate a charge ratio based on the production data and the consumption data, the charge ratio representing a percentage of energy produced by the energy asset that is available for charging a storage device;compare the consumption correlation value to a predetermined correlation threshold;compare the charge ratio to a predetermined charge ratio threshold; and determine that an issue associated with the current transformer has occurred when the consumption correlation value exceeds the predetermined correlation threshold or when the charge ratio is below the predetermined charge ratio threshold.

2. The current transformer issue detection system of claim 1, wherein the instructions further cause the processor to translate the consumption correlation value into a binary score indicative of whether the correlation meets the predetermined correlation threshold.

3. The current transformer issue detection system of claim 1, wherein the charge ratio is calculated by determining differences between the production data and the consumption dataAttorney Docket No. 21830-160618-USover a defined time window, and calculating an average value indicative of the percentage of energy produced by the energy asset that is available for charging a storage device;.

4. The current transformer issue detection system of claim 1, wherein the instructions further cause the processor to:determine whether consumption data is present prior to calculating the consumption correlation value and the charge ratio; andin response to the consumption data being absent, determine that an issue associated with the current transformer has occurred.

5. The current transformer issue detection system of claim 1, wherein the instructions further cause the processor to:compare a number of data points in the consumption data to a predetermined consumption threshold; anddetermine that an issue has occurred when the number of data points in the consumption data with values at or below the predetermined consumption threshold exceeds a predetermined count threshold.

6. The current transformer issue detection system of claim 1, wherein the instructions cause the processor to retrieve asset data via an application programming interface and refine the asset data into the consumption data and the production data.

7. The current transformer issue detection system of claim 1, wherein the instructions cause the processor to periodically monitor the production data and consumption data after installation of the current transformer and recalculate the consumption correlation value and the charge ratio.Attorney Docket No. 21830-160618-US8. The current transformer issue detection system of claim 1, wherein the instructions further cause the processor to generate a report including details of the determined issue associated with the current transformer, the report including a prioritized list of tasks to be performed on the current transformer based on a type of issue associated with the current transformer.

9. The current transformer issue detection system of claim 1, wherein an issue associated with the current transformer corresponds to an installation or configuration error indicative of inaccurate consumption readings for an energy asset having an energy storage device, the inaccurate readings being associated with improper charging or discharging of the energy storage device.

10. The current transformer issue detection system of claim 1, wherein the instructions further cause the processor to initiate a maintenance scheduling operation in response to determining that an issue associated with the current transformer has occurred.

11. A non-transitory machine readable medium storing instructions that, when executed, cause a processor to:receive production data and consumption data associated with an energy asset coupled to a current transformer;calculate a consumption correlation value between the production data and the consumption data, the consumption correlation value representing a similarity between the production data and the consumption data;calculate a charge ratio based on the production data and the consumption data, the charge ratio representing a percentage of energy produced by the energy asset that is available for charging a storage device;compare the consumption correlation value to a predetermined correlation threshold; compare the charge ratio to a predetermined charge ratio threshold; andAttorney Docket No. 21830-160618-USdetermine that an issue associated with the current transformer has occurred when the consumption correlation value exceeds the predetermined correlation threshold or when the charge ratio is below the predetermined charge ratio threshold.

12. The non-transitory machine readable medium of claim 11, wherein the instructions further cause the processor to translate the consumption correlation value into a binary score indicative of whether the correlation meets the predetermined correlation threshold.

13. The non-transitory machine readable medium of claim 11, wherein the instructions further cause the processor to:determine whether consumption data is present prior to calculating the consumption correlation value and the charge ratio; andin response to the consumption data being absent, determine that an issue associated with the current transformer has occurred.

14. The non-transitory machine readable medium of claim 11, wherein the instructions further cause the processor to:compare a number of data points in the consumption data to a predetermined consumption threshold; anddetermine that an issue has occurred when the number of data points in the consumption data with values at or below the predetermined consumption threshold exceeds a predetermined count threshold.

15. The non-transitory machine readable medium of claim 11, wherein the instructions further cause the processor to generate a report including details of the determined issue associated with the current transformer, the report including a prioritized list of tasks to be performed on the current transformer based on a type of issue associated with the current transformer.Attorney Docket No. 21830-160618-US16. A method comprising:receiving production data and consumption data associated with an energy asset coupled to a current transformer;calculating a consumption correlation value between the production data and the consumption data, the consumption correlation value representing a similarity between the production data and the consumption data;calculating a charge ratio based on the production data and the consumption data, the charge ratio representing a percentage of energy produced by the energy asset that is available for charging a storage device;comparing the consumption correlation value to a predetermined correlation threshold; comparing the charge ratio to a predetermined charge ratio threshold; and determine that an issue has occurred when the number of data points in the consumption data with values at or below the predetermined consumption threshold exceeds a predetermined count threshold.

17. The method of claim 16, further comprising translating the consumption correlation value into a binary score indicative of whether the correlation meets the predetermined correlation threshold.

18. The method of claim 16, further comprising:determining whether consumption data is present prior to calculating the consumption correlation value and the charge ratio; andin response to the consumption data being absent, determining that an issue associated with the current transformer has occurred.

19. The method of claim 16, further comprising:compare a number of data points in the consumption data to a predetermined consumption threshold; andAttorney Docket No. 21830-160618-USdetermine that an issue has occurred when the number of data points in the consumption data with values at or below a predefined consumption value threshold exceeds a predetermined count threshold.

20. The method of claim 16, further comprising generating a report including details of the determined issue associated with current transformer, the report including a prioritized list of tasks to be performed on the current transformer based on a type of issue associated with the current transformer.Attorney Docket No. 21830-160618-US

Images