A method for improving crosstalk between HPLC and micropower wireless dual-mode communications

Abstract

The present invention discloses a method for improving the crosstalk between HPLC and micro-power wireless dual-mode communication stations, comprising: S1. After the node of the user terminal device is powered on, the signal acquisition time windows of the two communication modes are aligned; S2. The acquired dual-mode signal parameters are combined with the preset station topology database to generate a feature vector of multi-dimensional features; S3. A training and generation random forest model is performed, the feature vector of the multi-dimensional features is input, the station attribution weight value is dynamically calculated, and the feature vector is nonlinearly weighted; S4. Based on weight comparison and station boundary node processing, the station attribution weight value is combined with a preset impedance shielding compensation algorithm to correct the measurement error, generate a unique station identification and lock the networking relationship, and perform station attribution determination. While retaining the wide coverage advantage of HPLC, the present invention utilizes the spatial attenuation characteristics of wireless communication to achieve accurate identification of station boundaries, which can improve the instant installation and acquisition stability of the user terminal device.

Description

Technical Field

[0001] The present invention relates to the technical field of device communication fusion technology, and in particular to a collaborative networking method, device and storage medium for improving crosstalk in a substation area based on HPLC (high-speed power line carrier) and micro-power wireless dual-mode communication. Background Art

[0002] In the intelligent management of low-voltage distribution network substations, HPLC (high-speed power line carrier) technology has become the mainstream communication method for node device networking due to its advantages such as no need for additional wiring and strong penetration. However, due to the high-frequency coupling characteristics of HPLC signals when transmitted through power lines (2-12MHz frequency band), capacitive coupling paths are easily formed between the low-voltage side lines of adjacent substation transformers, causing the carrier signal to propagate across substation boundaries.

[0003] In recent years, micropower wireless communication (470-510MHz frequency band) has gradually become an auxiliary communication means to form a dual-mode complementary solution with HPLC due to its characteristics of fast spatial attenuation and strong directionality. However, the following problems still exist in the traditional single HPLC networking process: (1) Cross-area misnetworking: Nodes in adjacent areas are mistakenly identified as members of the same area due to the high-frequency coupling effect; (2) Dynamic topology instability: In the installation and acquisition scenario, the newly added nodes will select the main node based on the instantaneous signal strength, resulting in repeated changes in the node ownership relationship in adjacent areas, seriously affecting the integrity of data acquisition; (3) Signal interference superposition: HPLC signals from multiple areas form standing wave interference in the coupled line, causing signal strength fluctuations, further exacerbating misjudgment.

[0004] In addition, although existing dual-mode communication solutions attempt to combine wireless communication, they have the following limitations.

[0005] (1) Failure of independent decision-making: The HPLC and wireless modules each make networking decisions independently, and no cross-protocol layer coordination mechanism is established. When the dual-mode signal strength is at a critical value, conflicts still occur.

[0006] (2) Boundary identification blind spot: The existing solution relies on a single signal strength threshold (such as RSSI) to determine the substation area, without considering the shielding effect of power line impedance mutation points (such as the metal shell of the meter box) on the wireless signal, resulting in an increase of more than 40% in the misjudgment rate of boundary nodes.

[0007] To this end, the present application proposes a method for improving the crosstalk between HPLC and micro-power wireless dual-mode communications to solve the above technical problems. Summary of the Invention

[0008] The main purpose of the present invention is to provide a method for improving the crosstalk between HPLC and micro-power wireless dual-mode communication. By constructing a collaborative networking mechanism for HPLC and micro-power wireless dual-mode communication, while retaining the wide coverage advantage of HPLC, the spatial attenuation characteristics of wireless communication are used to achieve accurate identification of the substation boundaries, which can improve the installation and use stability of terminal equipment, so as to solve the technical problem of cross-substation misnetworking caused by high-frequency signal crosstalk among power substation nodes proposed in the background technology.

[0009] The present invention adopts the following technical solutions to solve the above technical problems:

[0010] A method for improving crosstalk between HPLC and micropower wireless dual-mode communications, comprising:

[0011] S1. After a user terminal device (such as a smart meter) is powered on, the built-in HPLC module and micropower wireless module are simultaneously activated. A hardware clock synchronization mechanism ensures that the signal acquisition time windows of the two communication modes are aligned. When the node is first powered on or the topology is updated, the HPLC carrier signal characteristics and wireless signal spatial propagation parameters are synchronously acquired as dual-mode signal parameters.

[0012] S2. The obtained dual-mode signal parameters are combined with the preset area topology database to generate a multi-dimensional feature vector;

[0013] S3. A random forest model trained on historical networking data uses a machine learning model to input feature vectors of multi-dimensional features, dynamically calculates the substation area attribution weights, and performs nonlinear weighted calculations on the feature vectors.

[0014] S4. Based on the weight comparison and area boundary node processing, the area ownership weight value is combined with the preset impedance shielding compensation algorithm to correct the measurement error, in order to generate a unique area identification and lock the networking relationship, perform the area ownership determination;

[0015] During the operation of steps S1-S4, the following steps are also performed synchronously:

[0016] S5. Set up a dual-mode instruction timing stamp alignment mechanism to synchronize the execution of networking instructions received by nodes in heterogeneous channels;

[0017] S6. Adopt a local topology self-healing strategy and perform dynamic topology stability maintenance through periodic dual-mode heartbeat detection.

[0018] Preferably, the specific execution process in step S1 is as follows:

[0019] S11.HPLC signal acquisition: Acquire the time domain waveform of the carrier signal near the zero-crossing point of the power line frequency cycle (±5ms window) and extract the signal-to-noise ratio (SNR), signal propagation delay, and harmonic distortion rate;

[0020] S12. Wireless Signal Acquisition: Receive micropower wireless signals via a directional antenna array within the synchronization time window and measure received signal strength (RSSI), angle of arrival (AoA), and multipath fading characteristics.

[0021] S13. Noise Suppression: Adaptively filter HPLC signals to eliminate high-frequency noise on power lines (such as inverter interference); perform multipath elimination on wireless signals to improve spatial positioning accuracy.

[0022] Preferably, the feature vector of the multidimensional feature in step S2 includes:

[0023] HPLC feature group: signal propagation path loss (calculated based on the transformer impedance spectrum in the substation area), adjacent node hierarchical relationship (including the number of hops from concentrator to branch box to meter);

[0024] Wireless feature group: signal azimuth deviation (the angle between the node-master station connection line and the normal line of the substation boundary), shielding attenuation coefficient (the signal attenuation value caused by the metal casing is predicted based on the meter box material database);

[0025] Dynamic environmental parameters: real-time power frequency load fluctuation index (reflecting the time-varying nature of the power line channel), spatial obstacle movement status (detected by wireless signal Doppler frequency shift).

[0026] Preferably, the specific process of generating the feature vector of the multi-dimensional feature in step S2 includes:

[0027] (1) Based on the impedance characteristics of the transformer in the substation, an HPLC path loss attenuation model is established to generate the signal propagation path loss of the HPLC feature group. The HPLC path loss attenuation model is:

[0028]

[0029] in, Refers to the HPLC path loss attenuation model, In order to obtain the wire distance from the node to the concentrator through the topology map, is the frequency of HPLC path loss, is the frequency attenuation exponent, is the transformer coupling loss;

[0030] (2) Aiming at the electromagnetic shielding effect of the metal meter box, a compensation model is constructed to generate the shielding attenuation coefficient of the wireless feature group. The compensation model is:

[0031]

[0032]

[0033] in, is the wireless signal shielding attenuation, is the straight-line distance from the meter box casing to the wireless module antenna, is the wireless signal wavelength, The material compensation coefficient pre-calibrated by ray tracing simulation, is the actual measured wireless signal reception strength, The wireless signal receiving strength after shielding compensation correction;

[0034] (3) The power frequency load fluctuation index is introduced to quantify the channel time-varying index, which is used to reflect the impact of load switching on the channel in real time, so as to generate the real-time power frequency load fluctuation index in the dynamic environment parameters, which is:

[0035]

[0036] in, The first The effective value of the voltage in a power frequency cycle, is the voltage reference value;

[0037] when When , it is determined that the channel enters a transient process and the update of the weight decision model is suspended.

[0038] Preferably, the specific operation process of step S3 includes:

[0039] S31. Based on historical network data, train M decision trees to form an ensemble model as a random forest model;

[0040] S32. Performing feature normalization on the multidimensional feature vector generated in step S2, and using the processed feature vector as input to the random forest model, wherein the multidimensional feature vector contains 12 key parameters;

[0041] S33. The random forest model performs nonlinear weighting on the input feature vector, with:

[0042]

[0043]

[0044] in, is the HPLC weight value, The voting results of the decision tree are obtained. For the The decision output of each tree;

[0045] The HPLC weight value and the wireless weight value respectively represent the probability of the node belonging to the current station area through the two communication methods. They are adaptively adjusted according to the real-time channel quality: the wireless weight ratio is automatically increased, and vice versa, the HPLC weight ratio is increased. The calculation formula is:

[0046]

[0047] in, is the signal-to-noise ratio of the HPLC signal;

[0048] When the HPLC channel deteriorates, the wireless weight ratio will be automatically and gradually increased based on the real-time channel quality (for example, when the HPLC signal-to-noise ratio is lower than 40dB). Otherwise, the HPLC weight ratio will be increased to avoid misjudgment caused by single channel failure.

[0049] Preferably, the specific operation process of executing the station area ownership determination and networking relationship locking in step S4 includes:

[0050] S41. Set the dual-mode weight joint criterion and perform the weight comparison operation. If the HPLC weight With wireless weights are higher than the preset threshold (such as and ), determine that the node belongs to this area, if the HPLC weight With wireless weights The difference exceeds the specified critical range (such as ), then start the secondary verification;

[0051] S42. For nodes at the junction of the substations (such as and The spatial constraints are all in the range of 0.4-0.6, and the arrival angle verification of wireless communication is forced to be enabled. When the deviation between the signal direction angle and the main station antenna pointing angle is less than the specified threshold (usually 15 degrees), the ownership relationship is confirmed;

[0052] S43. After successful determination, write the encrypted area identification code into the node storage chip, bind the node MAC address to the area ID on the concentrator side, establish a mapping relationship between the MAC address and the area ID, and the locking period is not less than 24 hours.

[0053] Preferably, the dual-mode weight joint criterion in step S41 includes:

[0054]

[0055] in, is the judgment output, is the HPLC weight, is the wireless weight, Preset thresholds for HPLC weights, Preset thresholds for wireless weights;

[0056] When the criterion output is 1, the node is confirmed to belong to this station area;

[0057] When the criterion output is 0, the hierarchical decision-making mechanism is started:

[0058] (a) Level 1 verification: If , it is determined to be a boundary node and direction angle verification is performed;

[0059] (b) Secondary verification: If and , determined as the HPLC dominant node, ignoring the wireless weight;

[0060] (c) Level 3 verification: If the decisions are inconsistent for three consecutive times, a manual review process will be triggered.

[0061] Preferably, the space constraint condition in step S42 is:

[0062]

[0063] in, is the arrival angle of the wireless signal estimated by the MUSIC algorithm; is the normal direction angle of the area boundary;

[0064] Before enabling the arrival angle verification of wireless communication, the quaternion Kalman filter is used to smooth the direction angle, which is:

[0065]

[0066] in, for The quaternion state estimate at time t, for The quaternion state estimate at time t, for The Kalman gain matrix at time , for The original direction angle measurement value at the moment, is the partial derivative matrix from quaternion to orientation angle, which is used to linearize the nonlinear observation model.

[0067] Preferably, the execution of step S5 includes the following specific operations:

[0068] S51. Set up the instruction synchronization mechanism: The concentrator broadcasts the network confirmation instruction through HPLC to embed the wireless channel reservation mark. After the wireless channel receives the reservation mark, it delays Send confirmation messages containing the same timestamp to ensure that the time difference of instructions received by the node in the dual-mode channel is less than the specified time difference (usually 1ms), with:

[0069]

[0070] in, Time deviation calibrated for the PTP protocol, is the optical fiber transmission distance, is the speed of light;

[0071] S52. According to the preset area topology database in step S2, the wireless signal strength reported by the node is dynamically compensated in real time (such as metal meter box node compensation +10dB), and the compensated signal strength is used for subsequent topology updates. The compensation algorithm is:

[0072]

[0073] in, is the wireless signal receiving strength after compensation, is the actual measured wireless signal reception strength, is the wireless signal shielding attenuation, The material compensation coefficient pre-calibrated by ray tracing simulation, is the relative displacement change between the node and the meter box monitored by the IMU sensor, is the initial installation distance;

[0074] S53. For nodes that still fail to meet the threshold after three consecutive compensations, they are marked as suspected cross-station nodes, their networking requests are suspended, and the manual review process is initiated.

[0075] Preferably, the specific operation process of step S6 includes:

[0076] S61. Design a differentiated heartbeat interval strategy and perform periodic heartbeat detection: the node sends a heartbeat packet through HPLC and wireless dual-mode channels. The core node sends a heartbeat packet every 15 minutes. If the designated channel has no response for two consecutive times, the weight model is dynamically adjusted. The edge node sends a heartbeat packet every 30 minutes. If the designated channel has no response for three consecutive times, the weight model is dynamically adjusted.

[0077] S62. Topology self-healing: When a sudden change in node signal characteristics is detected, exceeding a specified threshold (e.g., a Wh drop of more than 30%), local topology reconstruction is automatically initiated, recalculating weights only for the affected nodes, rather than reconstructing the entire network.

[0078] S63. Hash chain technology is used to ensure the integrity of topology data. After each topology update, the concentrator and nodes synchronize and verify the hash value to prevent data tampering.

[0079] Preferably, the specific operation process of the local topology reconstruction in step S62 includes:

[0080] L1. Take the fault node as the center and the radius Nodes within meters are included in the reconstruction domain;

[0081] L2. Re-execute steps S3-S4 only for nodes in the reconstruction domain.

[0082] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0083] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0084] As can be seen from the above technical solution, the present invention provides a method for improving the crosstalk between HPLC and micropower wireless dual-mode communication. Compared with the existing technology, the present invention has the following advantages:

[0085] 1. The present invention sets a multi-dimensional feature fusion and dynamic weight allocation mechanism in the networking decision model, which can combine the HPLC signal-to-noise ratio, wireless signal arrival angle and topological level characteristics to adaptively adjust the dual-mode communication weights, thereby avoiding misjudgment caused by single channel failure and reducing the misjudgment rate of station area attribution.

[0086] 2. By setting up a hardware clock synchronization mechanism at the physical layer, the present invention can forcibly align the HPLC and wireless signal acquisition time windows, eliminate asynchronous sampling errors, and solve the problem of instantaneous fluctuations in HPLC signals caused by sudden load changes in traditional solutions. This significantly improves the comparability of dual-mode data and reduces signal strength measurement errors.

[0087] 3. By presetting the meter box material database on the concentrator side, the present invention can dynamically compensate for the shielding attenuation of wireless signals caused by obstacles such as metal casings, restore the true signal strength, and solve the boundary misjudgment problem caused by ignoring the shielding effect, thereby reducing the boundary node misjudgment rate.

[0088] 4. By setting spatial constraints on the wireless signal arrival angle and the boundary normal direction in the determination of the area boundary, the present invention can increase the node position verification dimension, suppress the limitations of the single criterion of signal strength, eliminate the influence of signal crosstalk between adjacent areas, and reduce the cross-area networking error rate.

[0089] 5. By setting up a local reconstruction mechanism for abnormal nodes in dynamic maintenance, the present invention can recalculate the weights of only 10% of the abnormal nodes, thereby reducing the reconstruction time of the entire network, shortening the system recovery time, reducing energy consumption and improving topology stability.

[0090] 6. The present invention ensures strict synchronization of heterogeneous channel instructions by setting up a precise time protocol synchronization mechanism in the transmission of networking instructions, thereby eliminating logical conflicts and avoiding abnormal node responses caused by instruction transmission delays, thereby improving the success rate of networking instruction execution.

[0091] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easy to understand through the following description. Of course, it is not necessary to achieve all of the above-mentioned advantages simultaneously in order to implement any product of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0093] Figure 1 It is a schematic diagram of the overall operation flow of the present invention. DETAILED DESCRIPTION

[0094] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0095] In the embodiment, see Figure 1 .

[0096] The present invention proposes a method for improving crosstalk between HPLC and micropower wireless dual-mode communications. This method replaces traditional fixed threshold criteria by constructing a three-dimensional feature vector of the HPLC signal signal-to-noise ratio (SNR), the wireless signal angle of arrival (AoA), and the topological level, and dynamically calculating the cell attribution weight (W value) through a machine learning model. Furthermore, by designing a time slot alignment mechanism for HPLC and wireless communication, dual-mode signal acquisition time windows are synchronized at the physical layer (deviation <100μs), eliminating signal strength measurement errors caused by asynchronous sampling. Furthermore, a wireless signal shielding attenuation prediction model is established by pre-setting a cell topology map and a meter box location database, dynamically compensating for the wireless signal strength measurements at boundary nodes (by up to 8-12dB). This collaborative networking mechanism for HPLC and micropower wireless dual-mode communications allows precise identification of cell boundaries by leveraging the spatial attenuation characteristics of wireless communication while retaining the wide coverage advantage of HPLC.

[0097] like Figure 1 As shown, the method specifically includes the following operation process:

[0098] S1. After the smart meter is powered on, the built-in HPLC module and micropower wireless module are started simultaneously. A hardware clock synchronization mechanism ensures that the signal acquisition time windows of the two communication modes are aligned. When the node is first powered on or the topology is updated, the HPLC carrier signal characteristics and the wireless signal spatial propagation parameters are synchronously acquired as dual-mode signal parameters.

[0099] The specific execution process in step S1 is as follows:

[0100] S11.HPLC signal acquisition: Acquire the carrier signal's time domain waveform within a ±5ms window around the zero-crossing point of the power line frequency cycle, and extract the signal-to-noise ratio (SNR), signal propagation delay, and harmonic distortion rate.

[0101] S12. Wireless Signal Acquisition: Receive micropower wireless signals via a directional antenna array within the synchronization time window and measure received signal strength (RSSI), angle of arrival (AoA), and multipath fading characteristics.

[0102] At this time, it is important to integrate a high-precision clock source (such as a TCXO temperature-compensated crystal oscillator) on the node circuit board, and use FPGA to realize the same-source drive of the clock signals of the HPLC module and the wireless module. The following conditions must be met:

[0103]

[0104] in, is the sampling timestamp of the HPLC module, is the sampling timestamp of the wireless module, is the system sampling period (typical value 20ms, corresponding to the zero crossing point of the power frequency cycle);

[0105] In actual projects, a clock calibration algorithm is used to dynamically compensate for crystal oscillator drift. Every 5 minutes, a time calibration pulse is sent down through the concentrator. The node calculates the local clock deviation and updates the crystal oscillator control voltage to ensure long-term synchronization accuracy better than ±10μs.

[0106] S13. Noise suppression: Adaptively filters HPLC signals to eliminate high-frequency noise on power lines (such as inverter interference); performs multipath elimination on wireless signals to improve spatial positioning accuracy;

[0107] For high-frequency noise on the power line (such as interference from inverters and LED driver power supplies), a second-order adaptive notch filter is designed, which includes:

[0108]

[0109] in, is the center frequency of the interference to be suppressed (dynamic detection range 150Hz-2kHz), is the extreme convergence factor (a value of 0.97 can take into account both convergence speed and stability), is the HPLC sampling interval (1ms, corresponding to 1kHz sampling rate), represents the complex frequency of the discrete system, The corresponding unit delay (i.e., the delay of one sampling period) is used to construct the difference equation of the filter;

[0110] In specific implementation, the interference frequency component is detected in real time through fast Fourier transform (FFT). When the noise power in a certain frequency band exceeds the threshold (such as -50dBm), the And update the filter coefficients to achieve dynamic noise suppression.

[0111] In addition, in a specific embodiment, in order to improve the wireless positioning accuracy, a multipath channel impulse response model can be further established, including:

[0112]

[0113] Solve for the optimal equalizer weights via least squares estimation:

[0114]

[0115] in, For the The attenuation coefficient of each path (related to the obstacle material), is the total number of multipaths, is the unit impulse function, indicating that the signal The delay characteristics on each path, For the The phase shift of a path represents the phase change caused by the path propagation distance and the characteristics of the reflecting surface. For the The delay of the path relative to the direct path, is the received signal vector (dimension N×1), is the channel matrix (dimension N×K, including delay cyclic shift construction), is the equalizer weight vector (dimension K×1), is the regularization parameter;

[0116] In the actual project implementation, a sliding window approach is used to update the channel estimate. A matrix inversion operation (accelerated by Cholesky decomposition) is performed every 256 symbols received to ensure real-time performance.

[0117] In summary, the hardware clock synchronization mechanism forces the dual-mode signal acquisition time windows to be aligned (deviation <100μs), eliminating the signal strength measurement error caused by asynchronous sampling in traditional solutions (such as the instantaneous fluctuation of ±8dB of HPLC signals during sudden load changes), making the dual-mode data comparable.

[0118] S2. Combine the obtained dual-mode signal parameters with the preset substation topology database to generate a feature vector with multi-dimensional features.

[0119] The feature vectors of multidimensional features include:

[0120] HPLC feature group: signal propagation path loss (calculated based on the transformer impedance spectrum in the substation area), adjacent node hierarchical relationship (including the number of hops from concentrator to branch box to meter);

[0121] Wireless feature group: signal azimuth deviation (the angle between the node-master station connection line and the normal line of the substation boundary), shielding attenuation coefficient (the signal attenuation value caused by the metal casing is predicted based on the meter box material database);

[0122] Dynamic environmental parameters: real-time power frequency load fluctuation index (reflecting the time-varying nature of the power line channel), spatial obstacle movement status (detected by wireless signal Doppler frequency shift).

[0123] Furthermore, the specific process of generating the feature vector of the multi-dimensional feature includes:

[0124] (1) Establish a GIS topology database for the substation area, store parameters such as the material (copper / aluminum), cross-sectional area, and insulation thickness of each section of conductor, and dynamically calculate value, and then based on the transformer impedance characteristics of the substation, an HPLC path loss attenuation model is established to generate the signal propagation path loss of the HPLC feature group. The HPLC path loss attenuation model is:

[0125]

[0126] in, Refers to the HPLC path loss attenuation model, To obtain the wire distance from the node to the concentrator through the topology map (obtained through the topology map), is the frequency of HPLC path loss, is the frequency attenuation index (the measured value is 1.8±0.3, which is related to the aging degree of the wire), is the transformer coupling loss (typical value 15dB on the 380V side);

[0127] (2) Aiming at the electromagnetic shielding effect of the metal meter box, a compensation model is constructed to generate the shielding attenuation coefficient of the wireless feature group. The compensation model is:

[0128]

[0129]

[0130] in, is the wireless signal shielding attenuation, is the straight-line distance from the meter box casing to the wireless module antenna, is the wireless signal wavelength (470MHz corresponds to 0.638m), The material compensation coefficient pre-calibrated by ray tracing simulation, is the actual measured wireless signal reception strength, The wireless signal receiving strength after shielding compensation correction;

[0131] In actual deployment, the concentrator maintains a database of meter box attributes (including material, size, and installation angle). After the node is powered on, the GPS positioning coordinates are matched to the database and compensation parameters are automatically loaded.

[0132] (3) The power frequency load fluctuation index is introduced to quantify the time-varying index of the channel, which is used to reflect the impact of load switching on the channel in real time, so as to generate the real-time power frequency load fluctuation index in the dynamic environment parameters. , used to reflect the impact of load switching on the channel in real time, including:

[0133]

[0134] in, The first The effective value of the voltage in a power frequency cycle, is the voltage reference value (nominal 220V, ±10% fluctuation allowed), is the total number of power frequency cycles;

[0135] when When , it is determined that the channel enters a transient process and the update of the weight decision model is suspended.

[0136] In summary, the meter box material database (with pre-set metal and plastic material attenuation parameters) is introduced to compensate for shielding in wireless signal measurements. For example, a metal meter box can attenuate a 470MHz wireless signal by up to 15-20dB. This compensation algorithm restores the true signal strength, resolving the boundary misjudgment problem caused by traditional solutions that ignore shielding effects.

[0137] S3. Based on the random forest model trained with historical networking data, the machine learning model inputs the feature vector of multi-dimensional features, dynamically calculates the substation attribution weight value, and performs nonlinear weighted calculation on the feature vector.

[0138] The specific operation process at this time includes:

[0139] S31. Based on historical network data, train M = 200 decision trees to form an ensemble model as a random forest model;

[0140] S32. Multidimensional feature vector generated in step S2 Perform feature normalization / standardization processing, and there is a processed feature vector for:

[0141]

[0142] in, The first The historical mean of the dimensional feature, For the The standard deviation of the dimensional feature is used to eliminate dimensional differences;

[0143] For example, the HPLC signal-to-noise ratio is μ=58dB and σ=12dB; the wireless arrival angle deviation is μ=8° and σ=5°;

[0144] After processing, it is used as the input of the random forest model, where the multidimensional feature vector contains a total of 12 key parameters;

[0145] S33. The random forest model performs nonlinear weighting on the input feature vector, with:

[0146]

[0147]

[0148] in, is the HPLC weight value, The voting results of the decision tree are obtained. For the The decision output of each tree;

[0149] The HPLC weight value and the wireless weight value respectively represent the probability of the node belonging to the current station area through the two communication methods. They are adaptively adjusted according to the real-time channel quality: the wireless weight ratio is automatically increased, and vice versa, the HPLC weight ratio is increased. The calculation formula is:

[0150]

[0151] in, is the signal-to-noise ratio of the HPLC signal;

[0152] When the HPLC channel deteriorates, the wireless weight ratio will be automatically and gradually increased based on the real-time channel quality (for example, when the HPLC signal-to-noise ratio is lower than 40dB). Otherwise, the HPLC weight ratio will be increased to avoid misjudgment caused by single channel failure.

[0153] In summary, dynamic weighting based on channel quality adaptation is used instead of traditional fixed weighting. For example, if strong interference (SNR < 35dB) is detected on the HPLC channel, the wireless weighting ratio is increased from the default 50% to 80%, thus avoiding decision errors caused by a single channel failure.

[0154] S4. Based on weight comparison and substation boundary node processing, the substation ownership weight value is combined with the preset impedance shielding compensation algorithm to correct the measurement error to generate a unique substation identification and lock the networking relationship, and execute substation ownership determination and networking relationship locking.

[0155] The specific operation process for determining the substation ownership and locking the network relationship at this time includes:

[0156] S41. Set the dual-mode weight joint criterion and perform the weight comparison operation. If the HPLC weight And wireless weight , determine whether the node belongs to this area, if , then start the secondary verification;

[0157] The dual-mode weight joint criterion includes:

[0158]

[0159] in, is the judgment output, is the HPLC weight, is the wireless weight, The default threshold for HPLC weight is 0.7, which can be adjusted according to the density of the station area. Preset threshold for wireless weight, the default value is 0.6, which is related to antenna gain;

[0160] When the criterion output is 1, the node is confirmed to belong to this station area;

[0161] When the criterion output is 0, the hierarchical decision-making mechanism is started:

[0162] (a) Level 1 verification: If , it is determined to be a boundary node and direction angle verification is performed;

[0163] (b) Secondary verification: If and , determined as the HPLC dominant node, ignoring the wireless weight;

[0164] (c) Level 3 verification: If the decisions are inconsistent for three consecutive times, a manual review process is triggered;

[0165] S42. and For nodes at the intersection of the substations in the range of 0.4-0.6, spatial constraints are imposed and the arrival angle verification of wireless communication is forced to be enabled. When the deviation between the signal direction angle and the main station antenna pointing angle is less than 15°, the ownership relationship is confirmed;

[0166] The spatial constraints are:

[0167]

[0168] in, is the arrival angle of the wireless signal estimated by the MUSIC algorithm (accuracy ±3°); is the normal direction angle of the station area boundary (obtained from the GIS map);

[0169] Before enabling the arrival angle verification of wireless communication, the quaternion Kalman filter is used to smooth the direction angle, which is:

[0170]

[0171] in, for The quaternion state estimate at time t, for The quaternion state estimate at time t, for The Kalman gain matrix at time , for The original direction angle measurement value at the moment, is the partial derivative matrix from quaternion to orientation angle, which is used to linearize the nonlinear observation model;

[0172] S43. After successful determination, write the AES-256 encrypted station identification code into the node storage chip (non-volatile memory) ,have:

[0173]

[0174] in, AES-256 encryption. is the master key of the concentrator (unique for each zone), is the station ID, is the timestamp;

[0175] On the concentrator side, the node MAC address is bound to the station area ID, and a mapping relationship between the MAC address and the station area ID is established. The locking period is not less than 24 hours.

[0176]

[0177] in, Represented as a regional mapping relationship, The lock period is 24 hours by default and can be extended to 72 hours.

[0178] In summary, by proposing a mandatory azimuth verification mechanism, we can add spatial position constraints to boundary nodes. By combining the relationship between the wireless signal azimuth angle (AoA) and the normal direction of the station area boundary, we can reduce the misjudgment rate from 12% to below 2%.

[0179] In addition, the following steps are further performed synchronously during the operation of steps S1-S4:

[0180] S5. Set up a dual-mode instruction timing stamp alignment mechanism to enable the nodes to synchronously execute the networking instructions received in the heterogeneous channels.

[0181] In addition, the following specific operations are included in the execution of step S5:

[0182] S51. Set up the instruction synchronization mechanism: The concentrator broadcasts the network confirmation instruction through HPLC to embed the wireless channel reservation mark. After the wireless channel receives the reservation mark, it delays Send confirmation messages containing the same timestamp to ensure that the time difference of instructions received by the node in the dual-mode channel is less than 1ms, with:

[0183]

[0184] in, Time deviation calibrated for PTP protocol (typical value <1μs), is the optical fiber transmission distance, is the speed of light;

[0185] S52. Based on the preset area topology database in step S2, the wireless signal strength reported by the node is dynamically compensated in real time (e.g., +10dB compensation for metal meter box nodes), and the compensated signal strength is used for subsequent topology updates. The compensation algorithm is:

[0186]

[0187] in, is the wireless signal receiving strength after compensation, is the actual measured wireless signal reception strength, is the wireless signal shielding attenuation, The material compensation coefficient pre-calibrated by ray tracing simulation, is the relative displacement change between the node and the meter box monitored by the IMU sensor, is the initial installation distance;

[0188] S53. For nodes that still cannot meet the threshold after three consecutive compensations , marked as a suspected cross-station node, suspending its networking request and starting the manual review process, namely:

[0189]

[0190] in, Compensate for data operations;

[0191] when When a node fails, the following operations are performed: the node MAC address is recorded in the concentrator blacklist; the node networking request is suspended for 48 hours; and the work order system is triggered to dispatch an on-site maintenance task.

[0192] In summary, the dual-mode instruction timestamp alignment mechanism can ensure that the instructions received by nodes in heterogeneous channels are strictly synchronized, avoiding logical conflicts caused by instruction transmission delays (such as HPLC instructions triggering node responses before wireless instructions).

[0193] S6. Adopt a local topology self-healing strategy and perform dynamic topology stability maintenance through periodic dual-mode heartbeat detection.

[0194] At this time, the specific operation process of step S6 includes:

[0195] S61. Design a differentiated heartbeat interval strategy and perform periodic heartbeat detection: the node sends a heartbeat packet through HPLC and wireless dual-mode channels. The core node sends a heartbeat packet every 15 minutes. If the designated channel has no response for two consecutive times, the weight model is dynamically adjusted. The edge node sends a heartbeat packet every 30 minutes. If the designated channel has no response for three consecutive times, the weight model is dynamically adjusted.

[0196] At this time, the formula for determining the heartbeat response state is:

[0197]

[0198] in, is the number of HPLC channel confirmations, Number of wireless channel confirmations

[0199] S62. Topology self-healing: When a sudden change in node signal characteristics is detected, exceeding a specified threshold (e.g., a Wh drop of more than 30%), local topology reconstruction is automatically initiated, recalculating weights only for the affected nodes, rather than reconstructing the entire network.

[0200] At this time, the specific operation process of the local topology reconstruction in step S62 includes:

[0201] L1. Take the fault node as the center and the radius Nodes within meters are included in the reconstruction domain;

[0202] L2. Re-execute steps S3-S4 only for the nodes in the reconstruction domain. The computation time at this time satisfies the following conditions:

[0203]

[0204] in, The calculation time for a single node (the measured value in the specific embodiment is 0.8ms), is the communication overhead baseline value (50ms).

[0205] S63. Hash chain technology is used to ensure the integrity of topological data, including:

[0206]

[0207] After each topology update, the concentrator and nodes synchronize and verify the hash value to prevent data tampering.

[0208] In summary, the local topology self-healing strategy can reduce the traditional full network reconstruction time from minutes to seconds. For example, recalculating the weights of only 10% of abnormal nodes can reduce system recovery time from 120 seconds to less than 5 seconds.

[0209] At the same time, through the above process, the misnetworking rate of the present invention can be reduced from 15%-30% of the traditional solution to below 2%, the node switching frequency is reduced by 90%, and the dynamic maintenance energy consumption is reduced by 40%, achieving a fundamental solution to the cross-station crosstalk problem in the power communication network.

[0210] In a specific embodiment, in order to verify the actual effect of the method of the present application, the following test example operations are designed, and the test scenarios cover typical urban areas and complex electromagnetic environment scenarios, including:

[0211] (1) Configure the test environment

[0212]

[0213] (2) Set up a test plan

[0214]

[0215] (3) Comparison of key performance indicators

[0216]

[0217] (4) Specific case analysis

[0218] (a) Event 1: Cross-zone misconfiguration scenario (Scenario A)

[0219] Traditional solution: During the peak electricity consumption period at 7:00 PM, the HPLC signal coupling is enhanced due to the activation of the air conditioner cluster, and 32 nodes continuously jump between three stations (the jump period is 8-15 seconds).

[0220] The solution of the present invention: through wireless direction angle constraint (Δφ<12°), only one boundary node triggers manual review, reducing the number of misnetworking incidents by 97%.

[0221] (b) Event 2: High-power interference scenario (Scenario B)

[0222] Traditional solution: At the moment the arc furnace started (t=14:23:11), the HPLC signal-to-noise ratio dropped sharply to 28dB, causing 146 nodes to go offline.

[0223] The solution of the present invention: automatically switches to wireless dominant mode (Wr weight increases to 82%), only 9 nodes need to be partially reconstructed, and the system recovery time is shortened from 183 seconds in the traditional solution to 7 seconds.

[0224] (5) Comparison of data results

[0225] Comparison of misconfiguration rates: Scenario A: 22.4% → 1.7%; Scenario B: 34.8% → 2.1%

[0226] Comparison of dynamic reconstruction time: traditional solution: 118s (A) / 203s (B); present invention: 4.2s (A) / 6.8s (B).

[0227] Therefore, through dual-scenario testing and verification, this application achieves over 90% performance improvements in key indicators such as misnetworking suppression, topology stability, and anti-interference capabilities, while also reducing energy consumption by 40%-45%, fully meeting the design goals. In particular, system reliability is significantly improved in complex industrial scenarios, making it valuable for large-scale promotion and application.

[0228] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0229] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0230] In another embodiment provided by the present application, a computer program product containing instructions is also provided, which, when executed on a computer, enables the computer to execute any of the methods for improving station crosstalk between HPLC and micropower wireless dual-mode communications in the above embodiments.

[0231] It is understandable that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts of the above method.

[0232] The embodiment of the present application further provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus.

[0233] Memory for storing computer programs;

[0234] The processor is used to implement the above-mentioned method for improving the crosstalk between the HPLC and micro-power wireless dual-mode communications when executing the program stored in the memory.

[0235] The communication bus mentioned in the above electronic device can be a peripheral component interconnect standard bus or an extended industry standard architecture bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0236] The communication interface is used for communication between the above electronic device and other devices.

[0237] The memory may include a random access memory, or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0238] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor, etc.; it can also be a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.

[0239] It should also be noted that electronic devices also include terminal devices, which can also be called terminals, user equipment, mobile stations, mobile terminals, etc. Terminal devices can be mobile phones, smart TVs, wearable devices, tablet computers, computers with wireless transceiver functions, virtual reality terminal devices, augmented reality terminal devices, wireless terminals in industrial control, wireless terminals in unmanned driving, wireless terminals in remote surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, etc. The embodiments of this application do not limit the specific technology and specific device form used by the terminal devices.

[0240] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive).

[0241] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0242] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. In addition, the meaning of "and / or" appearing in the full text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or solutions in which A and B are satisfied at the same time. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in this field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A method for improving crosstalk between HPLC and micropower wireless dual-mode communication, characterized in that: include: S1. After the user terminal device node is powered on, the built-in HPLC module and micropower wireless module are simultaneously activated. A hardware clock synchronization mechanism ensures that the signal acquisition time windows of the two communication modes are aligned. When the node is first powered on or the topology is updated, the HPLC carrier signal characteristics and wireless signal spatial propagation parameters are synchronously acquired as dual-mode signal parameters. S2. The obtained dual-mode signal parameters are combined with the preset area topology database to generate a multi-dimensional feature vector. The multi-dimensional feature vector includes HPLC feature group parameters, wireless feature group parameters, and dynamic environment parameters. S3. Based on historical networking data, a random forest model is trained, the feature vector of the multi-dimensional feature is input into the random forest model, and a nonlinear weighted calculation is performed to obtain the substation attribution weight value; S4. Perform a weight comparison operation to determine whether the node belongs to this area or start a hierarchical decision-making mechanism. For nodes at the junction of the area, spatial constraints are imposed, and the arrival angle verification mechanism of wireless communication is forced to be enabled to generate a unique area identifier and lock the networking relationship, and perform the area ownership determination; During the operation of steps S1-S4, the following steps are also performed synchronously: S5. Set up a dual-mode instruction timing stamp alignment mechanism to synchronize the execution of networking instructions received by nodes in heterogeneous channels; S6. Adopt a local topology self-healing strategy, perform periodic dual-mode heartbeat detection, and use hash chain technology to ensure the integrity of topology data.

2. The method for improving the crosstalk between HPLC and micropower wireless dual-mode communication according to claim 1, wherein: The feature vector of the multidimensional feature in step S2 includes: HPLC feature group: signal propagation path loss, adjacent node hierarchical relationship; Wireless feature group: signal direction angle deviation, shielding attenuation coefficient; Dynamic environmental parameters: real-time power frequency load fluctuation index, spatial obstacle movement status.

3. The method for improving the crosstalk between HPLC and micropower wireless dual-mode communication according to claim 1, wherein: The specific process of generating the feature vector of the multi-dimensional feature in step S2 includes: (1) Based on the impedance characteristics of the transformer in the substation, an HPLC path loss attenuation model is established to generate the signal propagation path loss of the HPLC feature group. The HPLC path loss attenuation model is: in, Refers to the HPLC path loss attenuation model, is the wire distance from the node to the concentrator obtained through the topology map, is the carrier frequency associated with the HPLC path loss, is the frequency attenuation exponent, is the transformer coupling loss; (2) Aiming at the electromagnetic shielding effect of the metal meter box, a compensation model is constructed to generate the shielding attenuation coefficient of the wireless feature group. The compensation model is: in, is the wireless signal shielding attenuation, is the straight-line distance from the meter box casing to the wireless module antenna, is the wireless signal wavelength, The material compensation coefficient pre-calibrated by ray tracing simulation, is the actual measured wireless signal reception strength, The wireless signal receiving strength after shielding compensation correction; (3) The power frequency load fluctuation index is introduced to quantify the channel time-varying index, which is used to reflect the impact of load switching on the channel in real time, so as to generate the real-time power frequency load fluctuation index in the dynamic environment parameters, which is: in, The first The effective value of the voltage in a power frequency cycle, is the voltage reference value; when When , it is determined that the channel enters a transient process and the update of the random forest model is suspended.

4. The method for improving the crosstalk between HPLC and micropower wireless dual-mode communication according to claim 1, wherein: The specific operation process of the S3 step includes: S31. Based on historical network data, train M decision trees to form an ensemble model as a random forest model; S32. Perform feature normalization on the multidimensional feature vector generated in step S2, and use the processed feature vector as input to the random forest model; S33. The random forest model performs nonlinear weighting on the input feature vector, with: in, is the HPLC weight value, The voting results of the decision tree are obtained. For the The decision output of each tree; Adaptively adjust the HPLC weight value according to the signal-to-noise ratio of the HPLC signal, and there is an adaptively adjusted weight The calculation formula is: in, is the signal-to-noise ratio of the HPLC signal; When the HPLC channel deteriorates, the wireless weight ratio is gradually increased to avoid misjudgment caused by single channel failure.

5. The method for improving the crosstalk between HPLC and micropower wireless dual-mode communication according to claim 4, wherein: The specific operation process of performing the station area ownership determination in step S4 includes: S41. Set the dual-mode weight joint criterion and perform the weight comparison operation. If the HPLC weight and the wireless weight are both higher than the preset threshold, the node is determined to belong to this station area. If the difference between the HPLC weight and the wireless weight exceeds the specified critical range, the hierarchical decision-making mechanism is activated; S42. Apply spatial constraints to nodes at the junction of the stations, forcing the activation of the arrival angle check for wireless communications. When the deviation between the signal direction angle and the main station antenna pointing direction is less than a specified threshold, the ownership relationship is confirmed; S43. After successful determination, the encrypted area identification code is written into the node storage chip, and the node MAC address is bound to the area ID on the concentrator side to establish a mapping relationship between the MAC address and the area ID.

6. The method for improving crosstalk between HPLC and micropower wireless dual-mode communication according to claim 5, wherein: The dual-mode weight joint criterion in step S41 includes: in, is the judgment output, is the HPLC weight, is the wireless weight, Preset thresholds for HPLC weights, Preset thresholds for wireless weights; When the criterion output is 1, the node is confirmed to belong to this station area; When the criterion output is 0, the hierarchical decision-making mechanism is started: (a) Level 1 verification: If , it is determined to be a boundary node and direction angle verification is performed; (b) Secondary verification: If and , determined as the HPLC dominant node, ignoring the wireless weight; (c) Level 3 verification: If the decisions are inconsistent for three consecutive times, a manual review process will be triggered.

7. The method for improving crosstalk between HPLC and micropower wireless dual-mode communication according to claim 5, wherein: The spatial constraints in step S42 are: in, is the arrival angle of the wireless signal estimated by the MUSIC algorithm; is the normal direction angle of the area boundary; Before enabling the arrival angle verification of wireless communication, the quaternion Kalman filter is used to smooth the direction angle, which is: in, for The quaternion state estimate at time t, for The quaternion state estimate at time t, for The Kalman gain matrix at time , for The original direction angle measurement value at the moment, is the partial derivative matrix from quaternion to orientation angle.

8. The method for improving crosstalk between HPLC and micropower wireless dual-mode communication according to claim 1, wherein: The execution of step S5 includes the following specific operations: S51. Set up the instruction synchronization mechanism: The concentrator broadcasts the network confirmation instruction through HPLC to embed the wireless channel reservation mark. After the wireless channel receives the reservation mark, it delays Send confirmation messages containing the same timestamp to ensure that the time difference of the instructions received by the node in the dual-mode channel is less than the specified time difference, with: in, Time deviation calibrated for the PTP protocol, is the optical fiber transmission distance, is the speed of light; S52. According to the preset area topology database in step S2, the wireless signal strength reported by the node is dynamically compensated in real time, and the compensated signal strength is used for subsequent topology updates. The compensation algorithm is: in, is the wireless signal receiving strength after compensation, is the actual measured wireless signal reception strength, is the wireless signal shielding attenuation, The material compensation coefficient pre-calibrated by ray tracing simulation, is the relative displacement change between the node and the meter box monitored by the IMU sensor, is the initial installation distance between the node and the meter box; S53. For nodes that still fail to meet the threshold after three consecutive compensations, they are marked as suspected cross-station nodes, their networking requests are suspended, and the manual review process is initiated.

9. The method for improving crosstalk between HPLC and micropower wireless dual-mode communication according to claim 1, wherein: The specific operation process of step S6 includes: S61. Design a differentiated heartbeat interval strategy and perform periodic heartbeat detection: Nodes send heartbeat packets via HPLC and wireless dual-mode channels. Core nodes send heartbeat packets every 15 minutes. If a designated channel fails to respond for two consecutive times, a dynamic adjustment of the random forest model is triggered. Edge nodes send heartbeat packets every 30 minutes. If a designated channel fails to respond for three consecutive times, a dynamic adjustment of the random forest model is triggered. S62. Topology self-healing: When a change in a node's signal characteristics is detected exceeding a specified threshold, local topology reconstruction is automatically initiated, recalculating the weights of only the affected nodes rather than reconstructing the entire network. S63. Hash chain technology is used to ensure the integrity of topology data. After each topology update, the concentrator and nodes synchronize and verify the hash value to prevent data tampering.

10. The method for improving crosstalk between HPLC and micropower wireless dual-mode communication according to claim 9, wherein: The specific operation process of the local topology reconstruction in step S62 includes: L1. Take the fault node as the center and the radius Nodes within meters are included in the reconstruction domain; L2. Re-execute steps S3-S4 only for nodes in the reconstruction domain.

Images