A power quality monitoring method, device, and medium

Abstract

The application relates to the technical field of electric variable monitoring, and discloses a power quality monitoring method, equipment and medium, which comprises the following steps: collecting a power grid voltage waveform and extracting a phase, generating a frequency fingerprint based on a phase increment sequence, taming a PI structure of a local real-time clock and performing group synchronization; executing a variable sampling rate strategy according to a power frequency phase, and performing analysis after an event trigger; under the condition that there is no external standard source, using a voltage step or a transformer area voltage regulation switching in the operation of a power distribution system as a natural excitation, performing online identification and updating on the gain and bias of a metering link; decoupling transmission of a steady-state index as a state flow and an event reconstruction waveform as an event flow, and using recoverable ordered fragmentation of the event flow to perform encapsulation and recombination. The application forms systematic improvement in time reliability, event integrity, index accuracy and transmission recoverability, and is suitable for continuous monitoring and operation and maintenance of power quality of a power distribution network.

Description

A method, equipment and medium for power quality monitoring Technical Field

[0001] This invention relates to the field of electrical variable monitoring technology, and in particular to a method, device and medium for monitoring power quality. Background Technology

[0002] Power quality in distribution networks is constantly affected by the combined effects of distributed generation grid connection, rapid load fluctuations, frequent line topology switching, and cellular IoT link jitter. However, traditional monitoring methods often rely on fixed sampling rates and external time sources to obtain reference time, and then use static windows to calculate steady-state indicators and harmonic parameters. These solutions often encounter three bottlenecks in the field: first, the unavailability or short-term loss of synchronization of external time signals makes cross-device data comparison difficult; second, fixed sampling windows and fixed reporting cycles cannot simultaneously ensure the accuracy of event waveform capture and steady-state statistics; and third, under congestion and jitter conditions, public network links are prone to data backlog, event data loss, and difficulty in orderly retransmission and recovery, resulting in the loss of critical fault information. Meanwhile, with the increasing refinement of low-voltage distribution area voltage regulation and the growing number of power electronic devices at the load edge, existing terminals based on single-cycle and passive uploading mechanisms are no longer sufficient to meet the engineering requirements of time synchronization consistency, event priority, and recoverable transmission. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention addresses three core issues specific to the complex scenarios of no external time synchronization, highly fluctuating operation, and jittery links: First, it provides a time base self-calibration and group synchronization method based on the "fingerprint" of the power grid frequency, establishing a cross-terminal comparable time reference in the absence of GNSS; Second, it proposes an event-priority adaptive sampling and phase-locked dynamic window mechanism, using a unified phase caliber to obtain steady-state indicators and instantly rotating the window to fully capture transient states when an event is triggered; Third, it constructs a jitter-resistant dual-channel message strategy, decoupling the steady-state indicators calculated at fixed periods from the event waveform into two links: a state stream and an event stream. Through coordinated constraints of ordered fragmentation, missing fragment location, selective retransmission, fixed timeout, and reassembly time limit, it ensures reliable, recoverable, and auditable transmission of event loads in unstable links.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a power quality monitoring method, comprising:

[0006] Collect the grid voltage waveform and extract the phase, generate a frequency fingerprint based on the phase increment sequence, discipline the local real-time clock with a PI structure and perform group synchronization through a cloud device;

[0007] A variable sampling rate strategy is implemented based on the power frequency phase, and the analysis is performed by rotating the window aligned with the full-cycle anchor point after the event is triggered;

[0008] In the absence of an external standard source, the gain and bias of the metering link are identified and updated online by using the voltage step or transformer area voltage regulation switching during the operation of the power distribution system as a natural excitation.

[0009] Steady-state metrics are decoupled from the state stream and event reconstruction waveforms are transmitted as event streams. The event streams are encapsulated and reassembled using recoverable ordered fragmentation.

[0010] As a preferred embodiment of the power quality monitoring method of the present invention, the step of generating a frequency fingerprint based on a phase increment sequence includes providing a parameterized expression of the instantaneous frequency within a time window, as follows:

[0011] ,

[0012] in, Instantaneous frequency; For nominal power frequency; It is a dimensionless frequency offset; The frequency slope; For time; Take the window time center as the center, ; This is the initial sampling time. This is the Nth sampling time.

[0013] The observation equation for the phase increment is obtained by integrating the frequency over time:

[0014] ,

[0015] in, For the first Segment phase increment, ; For phase; The adjacent sampling interval ; For two adjacent sampling times, For the k-th sampling or calculation node, This refers to the time point of the (k-1)th sampling or calculation node; The increment domain is set to a constant bias; To observe the noise, Pi is the mathematical constant of a circle.

[0016] As a preferred embodiment of the power quality monitoring method of the present invention, the step of taming the local real-time clock using a PI structure includes converting scalar observations into a linear regression form:

[0017]

[0018] ,

[0019] in, For observation vectors; For designing the matrix; The k-th row of matrix X; This is the noise vector; The vector of parameters to be estimated; The number of samples; This represents the phase increment of the Nth segment;

[0020] Subsequently, a robust objective function with regularization is introduced.

[0021] ,

[0022] in, For the first Observation weights; For Huber's loss These are the regularization coefficients; for The There are 1 component, where k is the index;

[0023] The normal equation update formula for iterative weighted least squares is given:

[0024] ,

[0025] in, It is a diagonal weight matrix. For the first Observation weights, For residuals, Let k be the iterative weight of the k-th observation in IRLS. Let be the residual of the k-th sample; Let this be the Huber influence function; when Time to take It is a regular matrix; This is a transpose.

[0026] As a preferred embodiment of the power quality monitoring method of the present invention, the step of taming the local real-time clock using a PI structure further includes,

[0027] The frequency offset estimate is mapped to the local PI discipline ring and operates only on the clock calibration register, providing a consistent time baseline for data timestamps. The specific formula is as follows:

[0028] ,

[0029] in, For the first Frequency offset estimate for each window; This is an error signal; Let be the error signal for the m-th window; To control the quantity; These are the coefficients of the proportional and integral terms; This is a discrete sequence of clock calibration registers.

[0030] As a preferred embodiment of the power quality monitoring method of the present invention, the implementation of the variable sampling rate strategy includes,

[0031] The phase density and baseline rate are combined into a time-domain intensity function, and sampling is triggered when the value reaches an integer, resulting in a non-uniform timestamp:

[0032] ,

[0033] ,

[0034] ,

[0035] in, For time intensity; It is the instantaneous phase; It is a constant; From The cumulative intensity of the rise; To be according to The generated sampling time sequence; Indices are positive integers; for Temporal intensity at any given moment; For a specific moment; The target sampling density is the instantaneous phase angle; This is the start timestamp for the integration process; The phase angle variable at the current analysis time; This indicates when the cumulative intensity first reaches an integer value. At that time, record that moment as the actual sampling moment. ; The target sampling density is per unit phase angle;

[0036] After the event is triggered by the fast indicator, the starting point of the analysis window is rotated to the target phase, and an integer cycle length is used to ensure phase consistency in subsequent frequency domain calculations:

[0037] ,

[0038] ,

[0039] ,

[0040] in, The moment the event is triggered; Phase at the trigger time; The angular frequency at the trigger time; For the target phase; To ensure that the starting point falls on an integer part of the target phase in the near future; This is the time shift amount; This is the starting point of the rotated window;

[0041] Define an analysis window that covers an integer number of cycles, and define the equivalent period using the mean angular frequency within the window, while keeping the window consistent with the phase:

[0042] ,

[0043] ,

[0044] ,

[0045] in, To analyze the window interval; The number of cycles contained in the window; The length of the window; The mean angular frequency within the window; The equivalent period; For voltage signal in time The instantaneous angular frequency at a given moment.

[0046] As a preferred embodiment of the power quality monitoring method described in this invention, the analysis using window rotation aligned with the full-cycle anchor point includes, since the sampling is at non-uniform times, reconstructing the sample points to a uniformly spaced grid to unify the calculation basis for subsequent RMS and harmonic estimation:

[0047] ,

[0048] in, For maximization operators; This is a column vector of optimization variables used to solve the reconstruction problem; This is the original non-uniform sample sequence; To unify the grid signal vector; To standardize the number of grid sample points; For the sampling operator, its kernel ; To rebuild the nuclear core; To unify the grid time step; D is the difference regularization operator; The regularization coefficient; It is a 2-norm;

[0049] A phase-synchronized cosine window is used, and energy normalization is performed to ensure compatibility with non-flat windows. Then, harmonic estimation is performed under the same window weight.

[0050] ,

[0051] ,

[0052] in, This is a phase synchronization window function; These are different window coefficients; To unify the grid Individual sample point values; is the phase corresponding to the sample point; RMS is the root mean square value based on window energy normalization.

[0053] Using the same window weight as RMS, least squares estimation is performed on the integer basis function coefficients to obtain the amplitude and phase parameters of the fundamental frequency and each harmonic:

[0054] ,

[0055] in, A diagonal weighted matrix that matches the window function; Phase synchronization window function; B is the harmonic basis function matrix; The largest harmonic order; The coefficient vector of the fundamental frequency and harmonics to be estimated; To unify the grid Phase angle corresponding to each sample point; This represents the DC basis function value of the nth sample point; It is the element-wise square root form of a diagonal matrix; Harmonic order; For the first The sample points at the th The values ​​of the first harmonic cosine basis function; For the first The sample points at the th The values ​​taken on the sinusoidal basis function of the first harmonic.

[0056] As a preferred embodiment of the power quality monitoring method described in this invention, the online identification and updating of the gain and bias of the metering link includes: locating the voltage step under phase synchronization conditions, taking representative values ​​only from the steady-state integer windows before and after the step, then using the voltage regulation step characteristic to limit the real step to the discrete dictionary, setting the previous steady-state real value as a latent variable, and explicitly parameterizing the sensitivity of the gain and bias to temperature and frequency.

[0057] The observation equation is formed by substituting the two points before and after the event into the calculation, and the environment term is explicitly separated by difference. The steady-state true voltage, the step quantity under quantization constraints, and the environment parameterized measurement coefficient vector are solved sequentially by fixing the remaining variables. The process is iterated until convergence to obtain a set of initial calibration parameters. The parameters are updated by recursive least squares when a new event arrives and projected back to the physical feasible range of gain and bias to form a measurement coefficient that is available in real time and subject to boundary constraints.

[0058] As a preferred embodiment of the power quality monitoring method of the present invention, the decoupled transmission includes: specifying that the terminal divides the reported data into two independent channels, a state stream and an event stream, on the acquisition side, with the state stream having a fixed period. Calculate and report steady-state indices; if the steady-state threshold is not met, only the minimum basis set is reported.

[0059] Meanwhile, after the event stream is triggered, the same event payload is divided into ordered fragments with EventID, fragment number, references before and after, and CRC. The receiver reassembles them in order and selectively retransmits fragments with NACK for missing fragments. The sender maintains a limited in-transit window and a concurrency limit.

[0060] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a power quality monitoring method.

[0061] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a power quality monitoring method.

[0062] The beneficial effects of this invention are as follows: This invention achieves time reference consistency between terminals without relying on external systems, making steady-state indicators and event waveforms comparable across devices and distribution areas. Through phase locking and dynamic windows, steady-state statistics and event capture are unified to the same phase reference, reducing estimation bias caused by frequency drift and window leakage. Furthermore, by separating state and event flows and clearly defining event priorities, concurrency limits, in-transit windows, fixed timeouts, and reassembly limits, this invention significantly reduces the probability of event loss under link jitter conditions, preventing state data from crowding out uplink bandwidth. Simultaneously, combined with online metering self-calibration strategies, it utilizes load step changes or natural excitation from distribution area voltage regulation to continuously identify and update gain and bias, maintaining long-term consistency. Overall, this invention provides a systematic improvement in timekeeping reliability, event integrity, indicator accuracy, and transmission recoverability, making it suitable for continuous monitoring and maintenance of power quality in distribution networks. Attached Figure Description

[0063] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 is a schematic flowchart of a power quality monitoring method according to an embodiment of the present invention. Detailed Implementation

[0065] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0066] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0067] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0068] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0069] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0070] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0071] Example 1, referring to Figure 1, is the first embodiment of the present invention. This embodiment provides a power quality monitoring method, including:

[0072] S1: Acquire the grid voltage waveform and extract the phase, generate a frequency fingerprint based on the phase increment sequence, tame the local real-time clock using a PI structure, and perform group synchronization through a cloud device.

[0073] The generation of frequency fingerprints based on phase increment sequences includes providing a parameterized expression for the instantaneous frequency within a time window, as shown below:

[0074] ,

[0075] in, Instantaneous frequency; For nominal power frequency; It is a dimensionless frequency offset; The frequency slope; For time; Take the window time center as the center, ; The initial sampling time, This is the time of the Nth sampling (half-cycle anchor point).

[0076] The observation equation for the phase increment is obtained by integrating the frequency over time:

[0077] ,

[0078] in, For the first Segment phase increment, ; For phase; For adjacent sampling intervals, ; For two adjacent sampling times, For the k-th sampling or calculation node, This refers to the time point of the (k-1)th sampling or calculation node; The increment domain is set to a constant bias; To observe the noise, Pi is the mathematical constant of a circle.

[0079] The taming of the local real-time clock using a PI structure includes transforming scalar observations into a linear regression form:

[0080] ,

[0081] ,

[0082] in, For observation vectors; For designing the matrix; The k-th row of matrix X; This is the noise vector; The vector of parameters to be estimated; The number of samples; This represents the phase increment of the Nth segment;

[0083] Subsequently, a robust objective function with regularization is introduced.

[0084] ,

[0085] in, For the first Observation weights; For Huber's loss These are the regularization coefficients; for The There are 1 component, where k is the index;

[0086] The normal equation update formula for iterative weighted least squares is given:

[0087] ,

[0088] in, It is a diagonal weight matrix. For the first Observation weights, For residuals, Let k be the iterative weight of the k-th observation in IRLS. Let be the residual of the k-th sample; Let this be the Huber influence function; when Time to take It is a regular matrix; This is a transpose.

[0089] The frequency offset estimate is mapped to the local PI discipline ring and operates only on the clock calibration register, providing a consistent time baseline for data timestamps. The specific formula is as follows:

[0090] ,

[0091] in, For the first Frequency offset estimate for each window; This is an error signal; Let be the error signal for the m-th window; To control the quantity; These are the coefficients of the proportional and integral terms; This is a discrete sequence of clock calibration registers.

[0092] S2: Executes a variable sampling rate strategy based on the power frequency phase, and performs analysis by rotating the window aligned with the full-cycle anchor point after the event is triggered.

[0093] The phase density and baseline rate are combined into a time-domain intensity function, and sampling is triggered when the value reaches an integer, resulting in a non-uniform timestamp:

[0094] ,

[0095] ,

[0096] ,

[0097] in, For time intensity; It is the instantaneous phase; It is a constant; From The cumulative intensity of the rise; To be according to The generated sampling time sequence; Indices are positive integers; for Temporal intensity at any given moment; For a specific moment; The target sampling density is the instantaneous phase angle; This is the start timestamp for the integration process; The phase angle variable at the current analysis time; This indicates when the cumulative intensity first reaches an integer value. At that time, record that moment as the actual sampling moment. ; The target sampling density is the unit phase angle.

[0098] After the event is triggered by the fast indicator, the starting point of the analysis window is rotated to the target phase, and an integer cycle length is used to ensure phase consistency in subsequent frequency domain calculations:

[0099] ,

[0100] ,

[0101] ,

[0102] in, The moment the event is triggered; Phase at the trigger time; The angular frequency at the trigger time; For the target phase; To ensure that the starting point falls on an integer part of the target phase in the near future; This is the time shift amount; This is the starting point of the rotated window.

[0103] Define an analysis window that covers an integer number of cycles, and define the equivalent period using the mean angular frequency within the window, while keeping the window consistent with the phase:

[0104] ,

[0105] ,

[0106] ,

[0107] in, To analyze the window interval; The number of cycles contained in the window; The length of the window; The mean angular frequency within the window; The equivalent period; For voltage signal in time The instantaneous angular frequency at a given moment.

[0108] The analysis using window rotation aligned with the full-cycle anchor point includes, since the sampling is non-uniform, reconstructing the sample points to a uniformly spaced grid to unify the calculation basis for subsequent RMS and harmonic estimations.

[0109] ,

[0110] in, For maximization operators; This is a column vector of optimization variables used to solve the reconstruction problem; This is the original non-uniform sample sequence; To unify the grid signal vector; To standardize the number of grid sample points; For the sampling operator, its kernel ; To rebuild the nuclear core; To unify the grid time step; D is the difference regularization operator; The regularization coefficient; It is a 2-norm.

[0111] A phase-synchronized cosine window is used, and energy normalization is performed to ensure compatibility with non-flat windows. Then, harmonic estimation is performed under the same window weight.

[0112] ,

[0113] ,

[0114] in, This is a phase synchronization window function; These are different window coefficients; To unify the grid Individual sample point values; is the phase corresponding to the sample point; RMS is the root mean square value based on window energy normalization.

[0115] Using the same window weight as RMS, least squares estimation is performed on the integer basis function coefficients to obtain the amplitude and phase parameters of the fundamental frequency and each harmonic:

[0116] ,

[0117] in, A diagonal weighted matrix that matches the window function; Phase synchronization window function; B is the harmonic basis function matrix; The largest harmonic order; The coefficient vector of the fundamental frequency and harmonics to be estimated; To unify the grid Phase angle corresponding to each sample point; This represents the DC basis function value of the nth sample point; It is the element-wise square root form of a diagonal matrix; Harmonic order; For the first The sample points at the th The values ​​of the first harmonic cosine basis function; For the first The sample points at the th The values ​​taken on the sinusoidal basis function of the first harmonic.

[0118] S3: In the absence of an external standard source, the gain and bias of the metering link are identified and updated online by using the voltage step or transformer area voltage regulation switching during the operation of the power distribution system as a natural excitation.

[0119] Under phase synchronization conditions, the voltage step is located, and representative values ​​are taken only from the steady-state integer windows before and after the step:

[0120] ,

[0121] when h triggers the event, retrieving the pre-step steady-state index set. With step-steady-state index set Weekly average:

[0122] ,

[0123] Only when and The event is determined to be stable and reliable. After confirming stability and reliability, the event is... Including estimates, among which .

[0124] in, For the first The absolute moment on the timeline at the starting point of each half-week; For indexing, It is the square of the instantaneous voltage. Time is in the form of an integral variable. For the defined first The root mean square voltage of one and a half cycles, For CUSUM cumulative statistics, For n-1 cumulative statistics of CUSUM, This represents the half-cycle RMS value in the case of no step jump. The long-term baseline mean, Here, h is the drift compensation constant in CUSUM, and h is the CUSUM threshold. For the event The representative value of the steady-state voltage measured by the device prior to the event; The cardinality of the set is the number of integer cycles contained within the steady-state window, and it is a positive integer. Index for the whole week For the first The root mean square voltage calculated within an integer window. For the event The device measures the representative value of the steady-state voltage after the event occurs; For set The base of is a positive integer; The observed step amplitude, The minimum threshold for observing the step amplitude. The minimum number of integer cycles for the steady-state window. Initialize the value for CUSUM.

[0125] By utilizing the voltage regulation step characteristic, the true step jump is confined to a discrete dictionary, and the pre-steady-state true value is set as a latent variable:

[0126] ,

[0127] Explicitly parameterize the sensitivity of gain and bias to temperature and frequency:

[0128] ,

[0129] Substitute the two points before and after the event into the formula to form the observation equation, and construct a difference equation to explicitly separate the environment term:

[0130] ,

[0131] ,

[0132] in, The observed step amplitude is obtained directly from the device readings; For the set of allowable actual voltage step amplitudes; This refers to the number of steps for the voltage regulator settings; The nominal stride ratio for a single gear; Rated voltage; These represent the minimum and maximum allowed number of steps per gear, respectively; It is a set of integers; For the event The actual voltage before the step jump; The initial voltage; For the event The actual voltage after the step jump; This refers to the voltage readings reported by the device itself and output by the metering link; The measured temperature; This refers to the instantaneous power frequency. This is the actual voltage; It is the bias function; It is the gain function; This is the reference value for the gain under the reference environment; This is the linear coefficient of gain with respect to temperature; This is the linear coefficient of gain with respect to frequency; For reference frequency; This is the reference value biased at the reference temperature; This is the linear coefficient of the bias with respect to temperature; For the device in the event Measurement noise at the steady-state point before the step jump; For the event Measurement noise at the steady-state point after a step jump; The difference in noise between the two steady-state points. ; For the event Temperature at the moment corresponding to the steady-state window before the step jump; For the event Temperature at the steady-state window following the step jump; For the event The frequency at the moment corresponding to the steady-state window before the step jump; For the event The frequency at the steady-state window following the step jump; This represents the environmental drift term introduced by the gain changing with the environment; This represents the bias temperature drift term introduced by the bias changing with temperature. This is a reference temperature.

[0133] By solving the pre-steady-state true voltage, the step quantity under quantization constraints, and the environmental parameterized measurement coefficient vector sequentially with the remaining variables fixed, and iterating until convergence, a set of initial calibration parameters is obtained. Then, recursive least squares is used to update the parameters when a new event occurs and project them back to the physical feasible range of gain and bias to form real-time available and boundary-constrained measurement coefficients.

[0134] S4: Decouple the steady-state index as a state stream and the event reconstruction waveform as an event stream for transmission, and encapsulate and reassemble the event stream using recoverable ordered fragmentation.

[0135] The decoupled transmission includes specifying that the terminal on the acquisition side divides the reported data into two independent channels: a state stream and an event stream, with the state stream transmitting data at a fixed period. (The window is phase-aligned and covers integer cycles) Calculate and report steady-state indices. If the steady-state threshold is not met, only the minimum basis set (including timestamp, RMS, frequency, and status bits) is reported.

[0136] Meanwhile, after the event stream is triggered, the same event payload is divided into ordered fragments with EventID, fragment number, references before and after, and CRC. The receiver reassembles them in order and selectively retransmits fragments with NACK for missing fragments. The sender maintains a limited in-transit window and a concurrency limit.

[0137] When the link is congested or jittered, rate limiting and pause operations are performed on the state stream and heartbeats are sent (event streams are not degraded), and transmission is managed with fixed timeouts and reassembly timeouts; all messages use a unified time caliber (window center time) and parameter echoing to maintain consistency with the cloud, and on-chip CRC16 and event-level CRC32 checksums and terminal or cloud rolling buffers are used to ensure recoverability and auditability.

[0138] Example 2

[0139] The second embodiment of the present invention differs from the first embodiment in that:

[0140] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0144] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0145] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for monitoring power quality, characterized in that: This includes acquiring grid voltage waveforms and extracting phases, generating frequency fingerprints based on phase increment sequences, taming the local real-time clock using a PI structure, and performing group synchronization via a cloud device. Based on the power frequency phase, a variable sampling rate strategy is implemented. After the event is triggered, the window rotation aligned with the full-cycle anchor point is used for analysis. Under the condition of no external standard source, the voltage step or transformer area voltage regulation switching in the operation of the power distribution system is used as a natural excitation to identify and update the gain and bias of the metering link online. The steady-state index is decoupled as a state stream and the event reconstruction waveform is decoupled as an event stream. The event stream is encapsulated and reassembled using recoverable ordered fragmentation.

2. The power quality monitoring method as described in claim 1, characterized in that: The generation of frequency fingerprints based on phase increment sequences includes providing a parameterized expression for the instantaneous frequency within a time window, as shown below: ,in, Instantaneous frequency; For nominal power frequency; It is a dimensionless frequency offset; The frequency slope; For time; Take the window time center as the center, ； The initial sampling time, For the Nth sampling time, the observation equation for the phase increment is obtained by integrating the frequency over time: ,in, For the first Segment phase increment, ； For phase; For adjacent sampling intervals, ； For two adjacent sampling times, For the k-th sampling or calculation node, This refers to the time point of the (k-1)th sampling or calculation node; The increment domain is set to a constant bias; To observe the noise, Pi is the mathematical constant of a circle.

3. The power quality monitoring method as described in claim 2, characterized in that: The taming of the local real-time clock using a PI structure includes transforming scalar observations into a linear regression form: ， ,in, For observation vectors; For designing the matrix; The k-th row of matrix X; This is the noise vector; The vector of parameters to be estimated; The number of samples; Let N be the phase increment of the Nth segment; then a robust objective function with regularization is introduced. ,in, For the first Observation weights; For Huber's loss These are the regularization coefficients; for The Each component; and the normal equation update formula for iterative weighted least squares is given: ,in, It is a diagonal weight matrix. For the first Observation weights, For residuals, Let k be the iterative weight of the k-th observation in IRLS. Let be the residual of the k-th sample; Let this be the Huber influence function; when Time to take It is a regular matrix; This is a transpose.

4. The power quality monitoring method as described in claim 3, characterized in that: The PI-structure discipline of the local real-time clock also includes docking the frequency offset estimate to the local PI discipline loop, which operates only on the clock calibration register to provide a consistent time baseline for the data timestamps. The specific formula is as follows: ,in, For the first Frequency offset estimate for each window; This is an error signal; Let be the error signal for the m-th window; To control the quantity; These are the coefficients of the proportional and integral terms; This is a discrete sequence of clock calibration registers.

5. The power quality monitoring method as described in claim 4, characterized in that: The variable sampling rate strategy includes synthesizing the phase density and baseline rate into a time-domain intensity function, triggering sampling when the accumulated value reaches an integer, and obtaining a non-uniform timestamp: ， ， ,in, For time intensity; It is the instantaneous phase; It is a constant; From The cumulative intensity of the rise; Cumulative intensity from The first time the integral reaches a positive integer value The actual sampling time corresponding to the time, that is, satisfying The moment; Indices are positive integers; for Temporal intensity at any given moment; For a specific moment; The target sampling density is the instantaneous phase angle; This is the start timestamp for the integration process; The phase angle variable at the current analysis time; The target sampling density is defined as a unit phase angle; after the event is triggered by the fast indicator, the starting point of the analysis window is rotated to the target phase, and an integer cycle length is used to ensure phase consistency in subsequent frequency domain calculations. ， ， ,in, The moment the event is triggered; Phase at the trigger time; The angular frequency at the trigger time; For the target phase; To ensure that the starting point falls on an integer part of the target phase in the near future; This is the time shift amount; The starting point of the rotated window is defined; an analysis window covering an integer number of cycles is defined, and the equivalent period is defined using the mean angular frequency within the window, while maintaining the window's consistency with the phase: ， ， ,in, To analyze the window interval; The number of cycles contained in the window; The length of the window; The mean angular frequency within the window; The equivalent period; For voltage signal in time The instantaneous angular frequency at a given moment.

6. The power quality monitoring method as described in claim 5, characterized in that: The analysis using window rotation aligned with the full-cycle anchor point includes, since the sampling is non-uniform, reconstructing the sample points to a uniformly spaced grid to unify the calculation basis for subsequent RMS and harmonic estimations. ,in, For maximization operators; This is a column vector of optimization variables used to solve the reconstruction problem; This is the original non-uniform sample sequence; To unify the grid signal vector; To standardize the number of grid sample points; For the sampling operator, its kernel ； To rebuild the nuclear core; To unify the grid time step; D is the difference regularization operator; The regularization coefficient; The L2 norm is used; a phase-synchronized cosine sum window is employed and energy normalization is performed to ensure compatibility with non-flat windows. Subsequently, harmonic estimation is performed under the same window weight. ， ,in, This is a phase synchronization window function; These are different window coefficients; To unify the grid Individual sample point values; The sample point corresponds to the phase; RMS is the root mean square value based on window energy normalization; under the same window weight as RMS, least squares estimation is performed on the integer basis function coefficients to obtain the amplitude and phase parameters of the fundamental frequency and each harmonic: ,in, B is the diagonal weighted matrix that matches the phase synchronization window function; B is the harmonic basis function matrix. The largest harmonic order; The coefficient vector of the fundamental frequency and harmonics to be estimated; To unify the grid Phase angle corresponding to each sample point; This represents the DC basis function value of the nth sample point; It is the element-wise square root form of a diagonal matrix; Harmonic order; For the first The sample points at the th The values ​​of the first harmonic cosine basis function; For the first The sample points at the th The values ​​taken on the sinusoidal basis function of the first harmonic.

7. The power quality monitoring method as described in claim 6, characterized in that: The online identification and updating of the gain and bias of the metering link includes: locating the voltage step under phase synchronization conditions, taking representative values ​​only from the steady-state integer windows before and after the step, then using the voltage regulation step characteristic to limit the true step to a discrete dictionary, setting the pre-steady-state true value as a latent variable, and explicitly parameterizing the sensitivity of gain and bias to temperature and frequency; substituting the two points before and after the event into the calculation to form the observation equation, constructing a differential explicit separation of the environment term, and iterating until convergence by sequentially solving the pre-steady-state true voltage, the step quantity under quantization constraints, and the environment parameterized metering coefficient vector under the condition of fixed residual variables to obtain a set of initial calibration parameters; and using recursive least squares to update the parameters when a new event arrives and project them back to the physically feasible range of gain and bias to form real-time usable and boundary-constrained metering coefficients.

8. The power quality monitoring method as described in claim 7, characterized in that: The decoupled transmission includes specifying that the terminal on the acquisition side divides the reported data into two independent channels: a state stream and an event stream, with the state stream transmitting data at a fixed period. The system calculates and reports steady-state metrics; if the steady-state threshold is not met, only the minimum base set is reported. Simultaneously, after the event stream is triggered, it divides the same event payload into ordered fragments with EventID, fragment number, references, and CRC. The receiver reassembles the fragments in order and selectively retransmits fragments with NACK for missing fragments. The sender maintains a limited in-transit window and a concurrency limit.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images