POWER VECTOR ANALYST WITH ZERO TRACKING AND TRACKING GATES FOR POWER GRID MONITORING

Abstract

A power monitoring system comprises one or more power vector analyzers and a power controller with one or more ports for receiving transient event data, comprising one or more current images and associated metadata for a transient event from the one or more power vector analyzers, and one or more processors configured to execute code to cause the one or more processors to convert the one or more current images from the one or more power vector analyzers and the associated metadata into one or more transient event vectors and to store the one or more transient event vectors in a vector database.

Description

REFERENCE TO RELATED REGISTRATIONS

[0001] This disclosure is a non-provisional application and claims precedence over U.S. Patent Application No. 63 / 675,220 entitled “POWER VECTOR ANALYZER WITH TRACKING NULL AND TRACKING GATES FOR POWER GRID MONITORING”, filed on July 24, 2024, the disclosures of which are incorporated herein by reference in their entirety. This application is also related to U.S. Patent Application No. 18 / 914,685, “POWER VECTOR ANALYZER”, filed on October 14, 2024, the disclosures of which are incorporated herein by reference in their entirety. AREA OF TECHNOLOGY

[0002] This disclosure relates to power grid monitoring, in particular test and measurement systems for power grid monitoring and especially a power vector analyzer (PVA). BACKGROUND

[0003] US patent application No. 18 / 914,685, filed on October 14, 2024, hereinafter referred to as "the application '685", the contents of which are hereby incorporated into this application by reference, discloses a power vector analyzer (PVA) device for analyzing and monitoring the power on a line.

[0004] Typically, a unit called a phase meter monitoring unit (PMU) monitors the power grid. PMUs measure the voltage, current, and frequency of the current flowing through them. The PMU collects time-synchronized data and helps detect and identify disturbances in the power grid, such as power outages, voltage dips, and other power quality problems. PMUs are installed at various points in the power grid to provide a comprehensive overview of the system's condition and enable real-time monitoring and control.

[0005] PMUs typically display a phasor diagram, which represents a voltage vector and a current vector phase angle for each of the three phases of a three-phase power site in the grid. The phasor diagrams, such as 10 in Fig. The values ​​shown are continuously updated as the phase angles change due to reactive loads. Similarly, the voltage and current display on the right shows the voltage for the three different phases as 12, 14, and 16. Range 18 in the current display indicates a transient event. However, the current direction of the transient, its magnitude, its apparent power, and how it oscillates between inductive and capacitive reactance at any given time are not readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 shows examples of diagrams displayed by a phasor measurement unit. Fig. Figure 2 shows an embodiment of a power grid that uses power vector analyzers (PVAs). Fig. Figure 3 shows a system diagram of an embodiment of a power control system comprising several PVAs distributed over a power grid, a tensor generator, and an artificial intelligence with a vector database. Fig. Figure 4 shows an embodiment of a user interface on a power vector analyzer (PVA) that displays the quiescent current for three-phase power. Fig. Figure 5 shows an embodiment of a user interface on a PVA during zeroing of the quiescent current. Fig. Figure 6 shows an embodiment of a user interface on a PVA that displays the quiescent current set to zero. Fig. Figure 7 shows an embodiment of a user interface on a PVA that displays measured power transients relative to a zeroed quiescent current. Fig. Figure 8 shows an embodiment of a user interface on a PVA that displays tracking limit test circuits for three phases of power. Fig. Figure 9 shows an embodiment of a user interface on a PVA that displays a small transient event that exceeds the thresholds and generates triggers. Fig. Figure 10 shows an embodiment of a user interface on a PVA that displays a transient event after applying zero tracking. DETAILED DESCRIPTION

[0006] The embodiments disclosed herein offer improved visibility and understanding of the characteristics of transient events in the power grid, such as complex dynamic bidirectional apparent power transients. The source of these events, referred to herein as transient events, can include one of numerous examples, such as large induction current surges, sudden switching on of megawatt charging stations, fallen power lines, exploding power transformers, lightning strikes, and so on. The embodiments provide a visual display that captures all aspects of the dynamic transient fluctuations in magnitude, phase angle, and direction of power over the duration of the transient event.

[0007] The embodiments described herein normalize and isolate the transient event by incorporating a novel tracking boundary test circuit and determining a zero-tracking normalization that removes the quiescent current from each power distribution unit (PVA) that sees the event at different GPS locations. This feature allows the system to ignore the quiescent current, which varies continuously at a slower rate than would be the case with a transient event. The quiescent current can differ at different times at different PVA locations. The power level at different monitoring points in the network can change at different times at different points in the network without being associated with a transient event. The embodiments normalize or set to zero the quiescent current at each normalized location.In this way, each PVA location that sees the dynamic transient event will see the transient event as it changes or affects the PVA as the transient event moves from its origin to various distant locations. This results in a multidimensional view of the event, which could be useful for locating the event's position when supported by an AI vector database containing many types of past events and known information about those events.

[0008] The embodiments described herein also create image and text arrays for the captured transient event from each PVA that recorded the event. These embodiments create a tensor representation of the image group to obtain a single AI embedding vector for the event data from all PVAs. The language describes the key features of the event's cause, its location, or other information and is linked to the event embedding. Over time, as the vector database grows, the AI ​​system can classify new events occurring in the network. The vector database can be located in the power distribution centers, transient substations, control stations, etc., or distributed throughout the system, with automatic updates occurring between locations.

[0009] Fig. Figure 2 shows an embodiment of a power grid comprising PVAs at strategic locations, which in one embodiment are identifiable by GPS coordinates using satellites, as shown in Figure 38. The PVAs may be located at more or fewer locations than those shown. The power grid may, for example, include power plants, as shown in Figure 20; power pylons, as shown in Figure 22; power transmission and / or distribution stations, as shown in Figure 26; alternative energy generation sites, such as wind farms, 36; solar parks (not shown); a power distribution center (PDC), as shown in Figure 28; and one or more control centers, as shown in Figure 30. PVAs may also be located at electricity consumption sites, including office and industrial buildings, as shown in Figure 32; and residential buildings, as shown in Figure 34. The power grid may cover large areas with multiple subnetworks or, in the implementation of the embodiments, comprise a single subnetwork.The PDC 28 and the control center 30, as well as all PVAs in the network, communicate via wireless, wired, or satellite communication. Each PVA can communicate only with the PDC and / or the control center, or it can communicate with each other. In one embodiment, the [missing information] is described in... Fig. The vector database shown in Figure 37 aggregates all information from all PVAs. This data can then be used for grid monitoring and the formulation of derivatives related to grid improvements, including, but not limited to, mandated artificial intelligence / machine learning (AI / ML) services. The AI / ML services can include predictive measures, including, but not limited to, predictive maintenance, load balancing / peak distribution, and anomaly trend identification.

[0010] Fig. Figure 3 shows a block diagram illustrating how the PVAs and the control center can communicate about transient events and how the control center can classify them. As in relation to Fig. As explained in section 2, several PVAs, such as 24, are located at different positions in the power grid. All PVAs communicate data on transient events back to the control center 30. Each PVA can see the same event and provides a multidimensional view that may be unique to its location. The term "power images" used here encompasses the images transmitted by the PVAs that show transient events. These images may include the tracking boundary test circuits described below, which help identify the presence and, to some extent, the magnitude of the transient events.

[0011] The tensor creator 33 has the task of taking the multiple PVA mock stream images and metadata and consolidating them into a single tensor space to create a single vector. The tensor creation process includes part of the vector creation process and may involve inserting the images from the database into a movie and consolidating the movie into a single vector by routing it through an AI embedding model 35 and its transformers. Alternatively, each image and text can be routed, for example, through an AI model such as CLIP (Contrastive Language-Image Pre-training) to create a vector for the image and text in the vector database 37. If the process creates a vector for each stream image from each PVA, the multiple vectors for a single event must be combined, for example, by summarizing or averaging for comparison.The end result is that the AI ​​embedding is trained and, if necessary, pre-trained once, allowing new data to be continuously added and distinguished without requiring retraining, as the network characteristics can change over time. New transient events are aggregated into a tensor and then embedded into a vector. This vector is then compared to vectors stored in the database to classify the event's properties. If no match is found, the system notifies an operator, who then performs a classification on the vector, which is stored in the database.

[0012] As explained above, the vector database can also contain vectors related to monitoring the power grid during operation, when no transient events occur. The control center can collect information from the power supply units (PVAs) regularly or during specific non-transient events, such as peak load times, high-load events, etc. The machine learning (ML) system can create vectors for these events with solutions such as load balancing, changes in power distribution, etc. If conditions arise during normal operation that require a response, the ML system can provide services to better manage power distribution.

[0013] Fig. Figure 4 shows an embodiment of a PVA. The PVA of the embodiments can display the quiescent current signals on the power lines in the network. Fig. Figure 4 shows a user interface 40 with controls on the left and the user interface 40 on the right. Points 42, 44, and 46 represent the "reference" apparent power signal for each of the three phases in a polar network, with each phase 120° out of phase with the other two phases. Lines 48, 50, and 52, and the points on their axes, have zero phase between voltage and current. Points "above" the lines have inductive reactance, and points "below" the line have capacitive reactance. Lines 50 and 52 are 120 degrees out of phase with line 48 and with each other to represent the different phases on the power lines. The term "reference apparent power" refers to the apparent power displayed before the quiescent current is removed, to distinguish it from the apparent power measured after the zeroing process.

[0014] Fig. Figure 5 shows an embodiment of the user interface during the zeroing process, which is explained in detail below. The zeroing process essentially removes the quiescent current from the apparent power measurements, thereby enabling a more accurate representation and measurement of transient power fluctuations. This process may involve the use of a quadrature synchronous detector (QSD), as described in detail in application 685, which is incorporated above in its entirety by reference. A QSD or similar device generates a zero vector that "sets" the apparent quiescent current to zero, so that only the apparent power is then displayed. The lines, such as 54, between each point, such as 42, and the center of diagram 56 represent the process of determining the zero vector, which the system adds to the apparent reference power to remove the quiescent current from future measurements. Fig. Figure 6 shows the resulting display where all three points have moved towards the center 56, indicating that the quiescent current has been set to zero and the apparent power appears to be zero for all three phases.

[0015] Fig. Figure 6 also shows a limit circle 58. The limit circle allows the user or control center to define a limit mask. This defines the limit for apparent power measurements, specifying how far the zero centers may be "away" from or above the limit before triggering a response in the PVA. By applying the zero point, the quiescent current of each line was "removed" from the display, thus improving the detection of transient events.

[0016] Fig. Figure 7 shows transient apparent power fluctuations relative to zero, with the transients shown for each power phase, e.g., 60 for one phase. One or more of the lines have a power level that exceeds the boundary mask. When the transient apparent power signals exceed the boundary mask 58, the PVA, in one embodiment, sends the current pattern of the transient event and the associated metadata to the control center. The user interface displays the message "Transient Detected." It should be noted that the current pattern sent to the control center represents a likely scenario. The control center would receive current patterns from multiple PVAs, allowing it to gain an overview of the transient event geographically, since PVAs at different locations may have different data depending on their location.

[0017] It is possible for the PVA itself to act as a local node in the AI ​​network, detecting the transient, creating the vector, searching a local copy of the vector database, and receiving the event classification. It would then forward the classification to other local nodes, including the control center, allowing all local nodes to update their local copies of the vector database.

[0018] The figures above show events with a "fixed" zero point. The zero point does not change with the quiescent current at the PV system location in the grid. The embodiments shown here use a "tracking" zero point, so that transient events can be isolated from the quiescent current. This requires the use of a "tracking" limit circuit.

[0019] In the diagrams above, the limit circuit does not follow the center of the quiescent current. As mentioned above, the user can define a limit circuit to test the recorded power signals and identify a dynamic, transient event. If the system is not zero-tracked, there are three limit circuits, one for each power phase, as shown in Fig. Figure 8 shows, for example, the tracking boundary circle 62 for point 42. The zero tracking can be defined separately for each of the three tracking boundary circles that form the tracking boundary masks. The quiescent current in each circle is the center of the mask circle.

[0020] Fig. Figure 9 shows the display after a small transient event has occurred on the power lines. In the embodiment shown here, the tracking limit mask was exceeded in all three phases. The transient event, which triggers the message to the central control unit, could only occur in one of the phases. To obtain the current pattern, the quiescent current in each circuit is subtracted from the total event. In this view, all three limit mask circuits move to the center of the vector display. Now, the power for the transient event is displayed as a deviation from the quiescent current, as shown in Fig. 10 shown.

[0021] According to embodiments of the disclosure, power grid monitoring systems (PVAs) with zero tracking and tracking gates represent a significant advancement in power grid management and analysis. Embodiments of the disclosure include a novel approach to monitoring complex dynamic transient events in power grids, employing a sophisticated mechanism of tracking gate limit test circuits and zero tracking normalization. Embodiments of the disclosure capture triggered transient power events from multiple PVAs, which observe the event and transmit the data to an AI embedding system for classifying the event's key features. The innovative design and functionality of the PVAs highlight a significant advancement in modernizing and improving power grid analysis and maintenance.

[0022] Aspects of the disclosure may be executed on specially designed hardware, firmware, digital signal processors, or a specially programmed general-purpose computer, including a processor that operates according to programmed instructions. The terms "controller" or "processor" as used herein are intended to include microprocessors, microcomputers, application-specific integrated circuits (ASICs), and dedicated hardware controllers. One or more aspects of the disclosure may be embodied in computer-readable data and computer-executable instructions, for example, in one or more program modules executed by one or more computers (including monitoring modules) or other devices. In general, program modules include routines, programs, objects, components, data structures, and so on.Computer-executable instructions are instructions that perform specific tasks or implement certain abstract data types when executed by a processor in a computer or other device. These instructions can be stored on a non-volatile, computer-readable medium, such as a hard disk, optical disk, removable storage medium, solid-state storage, or random access memory (RAM). As is known to those skilled in the art, the functionality of program modules can be combined or distributed in various ways as desired. Furthermore, the functionality can be embodied, wholly or partially, in firmware or hardware equivalents such as integrated circuits, FPGAs, and the like.Certain data structures can be used to implement one or more aspects of the disclosure more effectively, and such data structures are considered within the context of the computer-executable instructions and computer-usable data described here.

[0023] The disclosed aspects may, in some cases, be implemented in hardware, firmware, software, or any combination thereof. The disclosed aspects may also be implemented as instructions executed by or stored on one or more non-transitory computer-readable media that can be read and executed by one or more processors. Such instructions may be referred to as a computer program product. Computer-readable media, as described herein, means any media that a computer device can access. For example, and without limitation, computer-readable media may include computer storage media and communication media.

[0024] Computer storage media are all media that can be used to store computer-readable information. For example, and without limitation, computer storage media can include RAM, ROM, electrically erasable programmable read-only memory (EEPROM), flash memory or other storage technologies, compact disc read-only memory (CD-ROM), digital video disc (DVD) or other optical storage media, magnetic cartridges, magnetic tapes, magnetic disk storage or other magnetic storage devices, and any other volatile or non-volatile, removable or non-removable media implemented in any technology. Computer storage media exclude signals per se and transient forms of signal transmission.

[0025] Communication media are all media that can be used to transmit computer-readable information. Examples of communication media include coaxial cables, fiber optic cables, air, or any other media suitable for transmitting electrical, optical, radio frequency (RF), infrared, acoustic, or other signals. EXAMPLES

[0026] The following are illustrative examples of the disclosed technologies. An embodiment of the technologies may comprise one or more of the examples described below, as well as any combination thereof.

[0027] Example 1 is a power monitoring system comprising: one or more power vector analyzers; and a power controller comprising: one or more ports for receiving transient event data comprising one or more current images and associated metadata for a transient event from the one or more power vector analyzers; and one or more processors configured to execute code to cause the one or more processors to: convert the one or more current images from the one or more power vector analyzers and the associated metadata into one or more transient event vectors; and store the one or more transient event vectors in a vector database.

[0028] Example 2 is the current monitoring system from Example 1, wherein the one or more processors are further configured to execute code to cause the one or more processors to: search the vector database for vectors that match one of the one or more transient event vectors; if a match is found, use a classification from the match to classify the transient event; and if no match is found, add the vector to the vector database.

[0029] Example 3 is the current monitoring system from Example 1 or 2, wherein the code that causes the one or more processors to convert the one or more current images and associated metadata into one or more transient event vectors includes code that causes the one or more processors to create a transient event vector for each current image and associated metadata.

[0030] Example 4 is the current monitoring system from Example 3, wherein the one or more processors are further configured to execute code that causes the one or more processors to: combine the transient event vectors for each current picture relating to the transient event by averaging or summarizing the transient event vectors to create a combined transient event vector; and use the combined transient event vector to search the vector database.

[0031] Example 5 is the current monitoring system from one of Examples 1 to 4, wherein the code that causes the one or more processors to convert the one or more current images and the associated metadata into one or more transient event vectors includes code that causes the one or more processors to insert each of the received one or more current images into an image sequence and to combine the image sequence into a transient event vector.

[0032] Example 6 is the power monitoring system from one of Examples 1 to 5, wherein the power controller is located at a central location and the one or more power vector analyzers are distributed over a power grid.

[0033] Example 7 is the power monitoring system from one of Examples 1 to 6, wherein each of the one or more power vector analyzers is distributed over a power grid and each power vector analyzer contains the power controller, and wherein the one or more power vector analyzers communicate with other one or more power vector analyzers to update the vector database at each power vector analyzer.

[0034] Example 8 is the power monitoring system from any of Examples 1 to 7, wherein each power vector analyzer further comprises one or more processors configured to execute code to cause the one or more processors to define a boundary mask for each phase of power, which is displayed on a display of the power vector analyzer.

[0035] Example 9 is the current monitoring system from Example 8, wherein the one or more processors are further configured to execute code that causes the one or more processors to: receive a signal indicating that the power vector analyzer should set the quiescent current to zero for each phase of the power displayed on the power vector analyzer display; and display the apparent power for each phase of the power.

[0036] Example 10 is the current monitoring system from Example 9, wherein the one or more processors are further configured to: determine that a transient event has occurred because the power in one or more phases has exceeded a tracking mask limit for that phase; subtract the quiescent current of each power phase from the total event power; and display a picture of the power for the transient event in the center of the display, the transient event being shown as a deviation from the quiescent current, and the picture of the transient event becoming one of the one or more current pictures.

[0037] Example 11 is the power monitoring system from any of Examples 1 to 10, wherein the vector database is located in a central location and the one or more processors are further configured to execute code that causes the one or more processors to: monitor the operation of the power grid; and provide one or more artificial intelligence and machine learning services that include at least one of the following functions: predictive measurements, predictive maintenance, load balancing, peak load distribution, and identification of anomaly trends based on monitoring.

[0038] Example 12 is a method for monitoring power in a network, comprising: receiving one or more current images with associated metadata for a transient event in at least one of three phases of power from one or more power vector analyzers; converting the current images from the one or more power vector analyzers and the associated metadata into one or more transient event vectors; and storing the one or more transient event vectors in a vector database.

[0039] Example 13 is the procedure from Example 12, which further includes: searching the vector database for vectors that match one or more of the transient event vectors; and, if a match is found, using a classification from the match to classify the transient event.

[0040] Example 14 is the procedure from Example 12 or 13, wherein converting the one or more stream patterns and the associated metadata into one or more transient event vectors involves creating a transient event vector for each stream pattern and the associated metadata.

[0041] Example 15 is the procedure from Example 14, wherein converting the one or more stream images and the associated metadata into one or more transient event vectors comprises: combining the transient event vectors for each stream image relating to the transient event by averaging or summarizing the transient event vectors to create a combined transient event vector; and using the combined transient event vector to search the vector database.

[0042] Example 16 is the procedure from one of Examples 12 to 15, wherein the conversion of the one or more stream images and the associated metadata into one or more transient event vectors comprises: arranging each of the one or more received stream images in an image sequence and combining the image sequence into a transient event vector.

[0043] Example 17 is the method from any one of Examples 12 to 16, which further comprises: receiving a signal at the one or more power vector analyzers indicating that the one or more power vector analyzers should set the quiescent current to zero for each of the power phases displayed on a user interface of one or more power vector analyzers; and displaying the apparent power for each power phase.

[0044] Example 18 is the procedure from Example 17, which further comprises: determining that a transient event has occurred because the power in one or more phases has exceeded a tracking mask limit for that phase; subtracting the quiescent current of each phase from a total event power; and displaying a picture of the power for the transient event in the center of the display, with the transient event being shown as a deviation from the quiescent current and the picture of the transient event becoming one of the one or more current pictures.

[0045] Example 19 is the method from any of Examples 12 to 18, which further comprises: monitoring the operation of the power grid; and providing one or more artificial intelligence and machine learning services, including at least one of the following: predictive measurements, predictive maintenance, load balancing, peak load distribution and detection of anomaly trends based on monitoring.

[0046] All features disclosed in the description, including the claims, the abstract, and the drawings, as well as all steps in a disclosed method or process, may be combined in any combination, except for combinations in which at least some of these features and / or steps are mutually exclusive. Any feature disclosed in the description, including the claims, the abstract, and the drawings, may be replaced by alternative features that serve the same, an equivalent, or a similar purpose, unless expressly stated otherwise.

[0047] Furthermore, this written description refers to certain features. It is understood that the disclosure in this specification includes all possible combinations of these certain features. For example, if a particular feature is disclosed in connection with a particular aspect, this feature may also be used, to the extent possible, in connection with other aspects.

[0048] Where this application refers to a procedure with two or more defined steps or operations, the defined steps or operations may be carried out in any order or simultaneously, provided that the context does not preclude such possibilities.

[0049] Although certain aspects of the revelation have been presented and described for the purpose of illustration, it is understood that various modifications can be made without deviating from the spirit and scope of the revelation. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 63 / 675.220

[0001] US 18 / 914.685 [0001, 0003]

Claims

[1] A power monitoring system comprising the following: one or more power vector analyzers; and a power regulator, including: one or more ports for receiving transient event data, comprising one or more current images and associated metadata for a transient event from the one or more power vector analyzers; and one or more processors trained to execute code to cause the one or more processors to do the following: Converting one or more current patterns from one or more power vector analyzers and their associated metadata into one or more transient event vectors; and Storing one or more transient event vectors in a vector database. [2] The power monitoring system according to claim 1, wherein the one or more processors are further configured to execute code to cause the one or more processors to do the following: Searching the vector database for vectors that match one or more transient event vectors; In the case of a match, a classification from the match is used to classify the transient event; and If no match is found, add the vector to the vector database. [3] The current monitoring system according to 1 or 2, wherein the code that causes the one or more processors to convert the one or more current images and associated metadata into one or more transient event vectors includes code that causes the one or more processors to create a transient event vector for each current image and associated metadata. [4] The power monitoring system according to claim 3, wherein the one or more processors are further configured to execute code that causes the one or more processors to do the following: to combine the transient event vectors for each stream pattern relating to the transient event into a combined transient event vector by averaging or summarizing the transient event vectors; and to use the combined transient event vector to search the vector database. [5] The current monitoring system according to any one of claims 1 to 4, wherein the code that causes the one or more processors to convert the one or more current images and associated metadata into one or more transient event vectors comprises code that causes the one or more processors to insert each of the received one or more current images into an image sequence and to combine the image sequence into a transient event vector. [6] The power monitoring system according to any one of claims 1 to 5, wherein the power controller is located at a central location and the one or more power vector analyzers are distributed over a power network. [7] The power monitoring system according to any one of claims 1 to 6, wherein each of the one or more power vector analyzers is distributed over a power network and each power vector analyzer contains the power controller, and wherein the one or more power vector analyzers communicate with other one or more power vector analyzers to update the vector database at each power vector analyzer. [8] The power monitoring system according to any one of claims 1 to 7, wherein each power vector analyzer further comprises one or more processors configured to execute code to cause the one or more processors to define a boundary mask for each phase of power, which is displayed on a display of the power vector analyzer. [9] The power monitoring system according to claim 8, wherein the one or more processors are further configured to execute code that causes the one or more processors to do the following: Receiving a signal indicating that the power vector analyzer should set the quiescent current to zero for each power phase displayed on the power vector analyzer's display; and to display the apparent power for each power phase. [10] The power monitoring system according to claim 9, wherein the one or more processors are further configured as follows: Determine that a transient event has occurred because the performance in one or more phases has exceeded a tracking mask limit for that phase; Subtract the quiescent current of each power phase from the total power output; and Displaying an image of the power for the transient event in the center of the display, with the transient event being shown as a deviation from the quiescent current and the image of the transient event becoming one or more current images. [11] The power monitoring system according to any one of claims 1 to 10, wherein the vector database is located at a central location and the one or more processors are further configured to execute code that causes the one or more processors to do the following: Monitoring the operation of the power grid; and Providing one or more artificial intelligence and machine learning services that include at least one of the following functions: predictive measurements, predictive maintenance, load balancing, peak power distribution, and anomaly trend detection based on monitoring. [12] A method for monitoring the performance in a network, comprising the following: Receiving one or more current images with associated metadata for a transient event in at least one of three phases of the current from one or more power vector analyzers; Converting the current patterns from one or more power vector analyzers and the associated metadata into one or more transient event vectors; and Storing one or more transient event vectors in a vector database. [13] The method according to claim 12, which further comprises: Searching the vector database for vectors that match one or more transient event vectors; and If a match is found, a classification from the match is used to classify the transient event. [14] The method according to claim 12 or 13, wherein converting the one or more stream images and the associated metadata into one or more transient event vectors comprises creating a transient event vector for each stream image and the associated metadata. [15] The method according to claim 14, wherein converting the one or more stream patterns and the associated metadata into one or more transient event vectors comprises: Combining the transient event vectors for each stream image related to the transient event by averaging or summing the transient event vectors to create a combined transient event vector; and Using the combined transient event vector to search the vector database. [16] The method according to any one of claims 12 to 15, wherein the conversion of the one or more stream images and the associated metadata into one or more transient event vectors comprises: arranging each of the one or more received stream images in an image sequence and combining the image sequence into a transient event vector. [17] The method according to any one of claims 12 to 16, further comprising: Receiving a signal at the one or more power vector analyzers indicating that the one or more power vector analyzers should set the quiescent current for each of the power phases displayed on a user interface of one or more power vector analyzers to zero; and Display of apparent power for each performance phase. [18] The method of claim 17, which further comprises: Determine that a transient event has occurred because the performance in one or more phases has exceeded a tracking mask limit for that phase; Subtract the quiescent current of each phase from the total event power; and Displaying an image of the power for the transient event in the center of the display, with the transient event being shown as a deviation from the quiescent current and the image of the transient event becoming one of the one or more current images. [19] The method according to any one of claims 12 to 18, further comprising: Monitoring the operation of the power grid; and Providing one or more artificial intelligence and machine learning services that include at least one of the following functions: predictive measurements, predictive maintenance, load balancing, peak power distribution, and anomaly trend detection based on monitoring.

Images