Electric power operation error prevention method and system based on multi-dimensional physical field frequency domain feature separation, and storage medium
By separating the frequency domain features of multidimensional physical fields and verifying gravity acceleration, the problems of off-site cheating and attitude change in power system anti-misoperation technology are solved, and the accurate verification of equipment status and anti-misoperation functions are realized.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-28
- Publication Date
- 2026-03-13
AI Technical Summary
Existing power system anti-misoperation technologies cannot effectively prevent off-site cheating and misjudgment of equipment status, especially when the equipment attitude changes or the power is off, resulting in poor robustness.
A multi-dimensional physical field frequency domain feature separation method is adopted. Environmental magnetic induction data is obtained through high sampling frequency, and dynamic and static geomagnetic features of power frequency are separated. Adaptive verification is carried out in combination with gravitational acceleration to achieve dual verification of identity and physical field.
It achieves highly robust error-proof verification during equipment operation and maintenance, prevents photo cheating, adapts to changes in terminal posture, and ensures accurate judgment of equipment status.
Smart Images
Figure CN121657162A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of power safety management and control technology, and in particular to a method, system and storage medium for preventing power operation errors based on the separation of frequency domain features of multidimensional physical fields. Background Technology
[0002] In the operation and maintenance of power systems, preventing accidental entry into energized compartments and misoperation of equipment (referred to as "misoperation prevention") is a core requirement for ensuring personal safety and power grid stability. Traditional misoperation prevention measures mainly rely on mechanical locks, computer key systems, and GPS / BeiDou-based positioning technology.
[0003] However, existing technical solutions have significant drawbacks in practical applications. First, simple QR code scanning or GPS positioning cannot completely eliminate "off-site cheating" behavior. For example, workers may take photos of the QR codes and send them to people who are not on-site for remote scanning and unlocking, or use GPS simulation software to tamper with location information.
[0004] Secondly, existing anti-misoperation technologies based on magnetic field fingerprints are usually limited to verifying static magnetic fields or confirming whether the equipment is energized by detecting power frequency magnetic fields. However, these technologies have two serious problems: First, when the posture of the operator holding the terminal (such as horizontal, vertical, or tilted) is inconsistent with the posture when the fingerprint is entered, the measured magnetic field vector components will change drastically, leading to verification failure and poor practicality; second, when the equipment is under power outage maintenance, the power frequency magnetic field disappears, and the anti-misoperation logic based on liveness detection will fail, making it impossible to distinguish between "equipment power outage" and "personnel absence".
[0005] Therefore, there is an urgent need for a highly robust anti-misoperation technology that can adapt to both equipment operation and maintenance conditions and is not affected by the handheld terminal's grip posture. Summary of the Invention
[0006] This application provides a method, system, and storage medium for preventing errors in power operations based on the separation of frequency domain features of multidimensional physical fields, in order to solve the aforementioned problems existing in the prior art.
[0007] Firstly, this application provides a method for preventing errors in power operations based on the separation of frequency domain features of multidimensional physical fields, including:
[0008] Step S1: In response to the work task instruction, parse the work task instruction to determine the task attributes of the equipment to be operated and the corresponding preset benchmark features; the task attributes include operating status or maintenance status;
[0009] Step S2: Obtain the identification information of the target device, and when in an operation position matching the method of obtaining the identification information, obtain time-series environmental magnetic induction intensity data collected at a first sampling frequency; the first sampling frequency is greater than twice the power system frequency.
[0010] Step S3: Perform frequency domain signal separation on the environmental magnetic induction intensity data, and extract the power frequency dynamic component features characterizing the equipment operating status and the static geomagnetic vector features characterizing the environmental structure from the same set of data;
[0011] Step S4: Perform adaptive physical field verification based on the task attribute: If the task attribute is in operation, determine whether the equipment meets the conditions for energized operation based on the power frequency dynamic component characteristics; if the task attribute is in maintenance, acquire synchronously collected gravity acceleration data, use the gravity acceleration data and the static geomagnetic vector characteristics to extract rotationally invariant features relative to the gravity vector direction, and verify the matching degree between the rotationally invariant features and the static environmental fingerprint in the preset reference features.
[0012] Step S5: When the identity information verification is successful and the adaptive physical field verification is successful, an unlock command is generated.
[0013] Secondly, this application provides a power operation error prevention system based on multi-dimensional physical field frequency domain feature separation, including a task parsing module, a multi-dimensional data acquisition module, a feature separation and processing module, a dual-mode verification module, and an unlocking control module.
[0014] Thirdly, this application provides a computer-readable storage medium.
[0015] The beneficial effects of this application are as follows: by using frequency domain separation technology, power frequency dynamic features and static geomagnetic features are extracted simultaneously from the same sensor data source, realizing adaptive verification of both equipment operation and maintenance states; in particular, by introducing gravitational acceleration for coordinate system projection, rotation-invariant features are extracted, solving the problem of fingerprint matching failure caused by changes in the posture of the handheld terminal; combined with dual verification of identity and physical field, it effectively defends against violations such as photo cheating. Attached Figure Description
[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below.
[0017] Figure 1 A flowchart illustrating the method provided in this application embodiment;
[0018] Figure 2 This is a schematic diagram illustrating the principle of frequency domain signal separation in the embodiments of this application;
[0019] Figure 3 This is a schematic diagram of the geometric principle of rotation-invariant feature extraction in the embodiments of this application;
[0020] Figure 4 This is a system structure block diagram provided for an embodiment of this application. Detailed Implementation
[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the method of recording the reference feature should be consistent with the operation method during on-site verification. For example, when using Near Field Communication (NFC) for identity recognition, the reference magnetic field feature should be data recorded close to the device surface; when using image recognition, the reference magnetic field feature should be data recorded at a standard shooting distance (e.g., 50cm).
[0022] Example 1: Image Recognition-Based Prevention of Misoperation Status of Live Equipment
[0023] This embodiment is applied to a substation inspection scenario, and the target device is a 110kV main transformer that is in a live operating state.
[0024] 1. Task Analysis: The operator receives the work order via a handheld smart terminal (such as an explosion-proof mobile phone). The system analyzes the work order, determines the current task attribute as "operational status inspection," and automatically retrieves the reference electromagnetic characteristics of the transformer in operation (such as a preset power frequency magnetic field activity threshold of 5 microtesla).
[0025] 2. Identification and Data Acquisition: The operator stands at a safe distance of approximately 1 meter from the equipment and uses the terminal camera to scan the QR code on the equipment (to acquire the identification image data). Simultaneously, the terminal's built-in magnetometer collects a 1-second sequence of environmental magnetic field strength data at a sampling rate of 200Hz (the first sampling frequency). Since the power system's operating frequency is 50Hz, according to the Nyquist sampling theorem, a sampling rate of 200Hz is greater than twice 50Hz, satisfying the requirement for distortion-free sampling.
[0026] 3. Frequency domain signal separation: The terminal performs a Fast Fourier Transform (FFT) on the acquired discrete magnetic flux density sequence B(t).
[0027] The background geomagnetic vector (DC component) is extracted by low-pass filtering (cutoff frequency set to 5Hz) or by calculating the mean.
[0028] The amplitude component at 50Hz is extracted from the spectrum to obtain the power frequency dynamic component characteristic B_50Hz.
[0029] 4. Adaptive Verification: Since the task attribute is "Running Status", the system enters the first verification branch. The system determines whether the calculated B_50Hz is greater than the preset active threshold (5 microtesla).
[0030] If B_50Hz = 12 microtesla > 5 microtesla, and the QR code is correctly parsed, it is determined that the operator was indeed near the live equipment and did not cheat by "taking a photo" (the photo has no power frequency magnetic field). The system generates an unlock command, allowing the filling in of the inspection record.
[0031] Example 2: Preventing Misoperation During Power Outage Maintenance Based on NFC and Rotation Invariant Features
[0032] This embodiment applies to a switchgear maintenance scenario where the target equipment is under power outage maintenance. In this scenario, there is no power frequency magnetic field, and the operator's handheld terminal posture (horizontal or vertical) is not fixed.
[0033] 1. Task Analysis: The system analyzes the work order, determines the task attribute as "power outage maintenance," and retrieves the static environmental fingerprint of the switchgear under power outage conditions. This fingerprint contains two core parameters: the vertical component reference value B_v_ref and the horizontal modulus reference value B_h_ref.
[0034] 2. Identity and Data Acquisition: The operator places the handheld terminal against the NFC tag on the device surface to read (acquire identity information). Due to the use of NFC, the operating distance is close to 0 meters. The terminal simultaneously collects magnetic field data B_raw = (Bx,By, Bz) and gravitational acceleration data G = (Gx, Gy, Gz).
[0035] 3. Feature separation and gravity compensation (rotation-invariant feature extraction):
[0036] Separation: Frequency domain analysis was also performed to confirm that the power frequency component B_50Hz was below the power outage noise threshold (e.g., 0.5 microtesla), thus verifying that the equipment had indeed lost power.
[0037] Gravity projection: The unit vector of gravity, g_hat = G / |G|, is determined using the gravitational acceleration G.
[0038] Calculate the vertical feature component (dot product): B_vertical = |B_raw · g_hat|.
[0039] Calculate the horizontal eigenvalue (Pythagorean theorem): B_horizontal_mod = sqrt(|B_raw|^2 - B_vertical^2).
[0040] Effect: Regardless of the angle at which the operator holds the terminal (i.e., how the x, y, z components in the device coordinate system change), as long as the position remains unchanged, the calculated B_vertical and B_horizontal_mod will be constant values.
[0041] 4. Matching Verification: The real-time calculated B_vertical and B_horizontal_mod are compared with B_v_ref and B_h_ref in the fingerprint database. If the error between the two is within ±10%, and NFC verification is successful, the physical location is determined to be correct and the device status matches, and an unlock command is generated.
[0042] Example 3: Anti-misplacement operation and auxiliary verification logic
[0043] This embodiment demonstrates robust handling in complex electromagnetic environments and security management after unlocking.
[0044] 1. Auxiliary Verification Logic (for special cases of operational status): Assuming that in Example 1, although the device is running, the power frequency magnetic field is weak due to light load (B_50Hz < threshold). In this case, the system does not directly report an error, but automatically switches to auxiliary verification mode: using the method in Example 2, it extracts the rotational invariant features of the static geomagnetic field and compares them with the reference fingerprint. If the static fingerprint matches successfully, and the system confirms that the device is indeed running despite the low current (confirmed through the background SCADA status), unlocking is still allowed, but an advanced warning "Please note that the device is powered on" will pop up.
[0045] 2. Relative Space Fence (After Unlocking): At time T0 when the unlock command is generated, the system locks the current position as the inertial origin (0,0,0). The terminal's built-in IMU (Inertial Measurement Unit) begins to collect acceleration and angular velocity at 100Hz, and performs short-term (e.g., within 30 seconds) double integration to calculate displacement.
[0046] Scenario: If, after unlocking, the operator walks with the terminal to an adjacent interval (more than 2 meters from the origin), the system calculates that the real-time displacement D > 2 meters (safe radius).
[0047] Action: The system immediately triggers the interlocking logic, forcibly exits the operation interface and alarms to prevent operators from violating regulations by "unlocking at point A and then operating at point B".
[0048] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A method for preventing errors in power operations based on the separation of frequency domain features of multidimensional physical fields, characterized in that, Includes the following steps: Step S1: In response to the work task instruction, parse the work task instruction to determine the task attributes of the equipment to be operated and the corresponding preset benchmark features; the task attributes include operating status or maintenance status; Step S2: Obtain the identification information of the target device, and when in an operation position matching the method of obtaining the identification information, obtain time-series environmental magnetic induction intensity data collected at a first sampling frequency; the first sampling frequency is greater than twice the power system frequency. Step S3: Perform frequency domain signal separation on the environmental magnetic induction intensity data, and extract the power frequency dynamic component features characterizing the equipment operating status and the static geomagnetic vector features characterizing the environmental structure from the same set of data; Step S4: Perform adaptive physics field verification based on the task attributes: If the task attribute is a running state, determine whether the equipment meets the conditions for energized operation based on the power frequency dynamic component characteristics. If the task attribute is maintenance status, acquire synchronously collected gravity acceleration data, use the gravity acceleration data and the static geomagnetic vector features to extract rotationally invariant features relative to the gravity vector direction, and verify the matching degree of the rotationally invariant features with the static environment fingerprint in the preset benchmark features. Step S5: When the identity information verification is successful and the adaptive physical field verification is successful, an unlock command is generated.
2. The method according to claim 1, characterized in that, The identification information of the target device is derived from at least one of image recognition data, near-field communication (NFC) data, radio frequency identification (RFID) data, or Bluetooth beacon data; the preset reference feature is pre-entered at a standard operating distance corresponding to the above acquisition method; Step S3 specifically includes: The obtained discrete magnetic induction intensity sampling sequence is low-pass filtered or smoothed to extract the original static magnetic field vector; The DC component in the sampling sequence is removed to obtain the dynamic sequence, and the dynamic sequence is subjected to spectral analysis to extract the amplitude at the corresponding power system frequency, which is used as the power frequency dynamic component feature.
3. The method according to claim 1, characterized in that, The step S4, "extracting rotationally invariant features relative to the direction of the gravity vector" and the verification process, specifically includes: The direction of the gravity vector is determined based on the aforementioned gravitational acceleration data; Calculate the projection component of the static geomagnetic vector feature in the direction of the gravity vector, and use it as the vertical feature component; Calculate the magnitude of the projection vector of the static geomagnetic vector feature onto a plane perpendicular to the direction of the gravity vector, and use it as the horizontal feature component; The vertical feature components and the horizontal feature components are compared with the corresponding reference components in the static environment fingerprint, respectively. When the deviations of both are within the preset tolerance range, the physical field verification under maintenance conditions is deemed to have passed.
4. The method according to claim 3, characterized in that, The verification process also includes: Determine whether the power frequency dynamic component characteristics are less than a preset power outage noise threshold; When the power frequency dynamic component characteristic is less than the power outage noise threshold and the rotation invariance characteristic passes the verification, the unlocking command is generated; The static environmental fingerprint is pre-recorded at the target device location while the device is powered off, and includes data on standardized vertical magnetic field components and horizontal magnetic field magnitudes.
5. The method according to claim 1, characterized in that, If step S4 is in the running state, the verification process specifically includes: Determine whether the power frequency dynamic component characteristics are greater than a preset activity threshold; If so, the physical field verification in the running state is deemed to have passed; If not, a prompt message will be output to indicate that the collection position should be adjusted, or the system will automatically switch to the auxiliary verification logic based on static environment fingerprints; if the auxiliary verification logic still fails, the locked state will remain. The active threshold is set based on the level of background electromagnetic noise in the environment.
6. The method according to claim 1, characterized in that, The method also includes anti-displacement operation steps based on relative spatial fencing: At the moment the unlock command is generated in step S5, a relative inertial coordinate system is established with the current spatial position as the origin; Using high-frequency motion data collected by the inertial measurement unit, the real-time displacement trend relative to the origin is calculated within a preset inertial estimation window. If the real-time displacement trend indicates that the preset safe operating radius has been exceeded, the interlocking logic is triggered.
7. A power operation error prevention system based on multi-dimensional physical field frequency domain feature separation, characterized in that, include: The task parsing module is used to respond to job instructions, parse task attributes, and retrieve corresponding baseline feature parameters; The multidimensional data acquisition module is used to acquire the target device's identification information, as well as magnetic induction intensity data and gravitational acceleration data that satisfy the sampling theorem; The feature separation and processing module is used to perform frequency domain separation on magnetic field data to obtain the dynamic component features of power frequency, and to extract the rotational invariant features of static magnetic field by combining gravitational acceleration data. The dual-mode verification module is configured to execute branch logic based on task attributes: when the task attribute is in a running state, the power frequency dynamic component characteristics are verified; when the task attribute is in a maintenance state, the rotational invariance characteristics are verified. The unlock control module is used to perform the unlock operation when the verification is successful.
8. A computer-readable storage medium having a computer program stored thereon, the computer program implementing the method as described in any one of claims 1 to 6 when executed by a processor.