A method and system for detecting multi-point electrostatic distribution of a pipeline

Abstract

The application relates to a pipeline multi-point static electricity distribution detection method and system. The method comprises the following steps: collecting pipeline multi-point static electricity potential data sequences and environmental parameter data sequences; performing decoupling processing on the static electricity potential data sequences, removing channel crosstalk interference to obtain decoupled potential data; calculating an environmental drift compensation amount based on a temperature and humidity deviation amount and a sensitivity coefficient, and correcting to obtain a compensated potential data sequence; discretely modeling an inner wall of a pipeline, constructing a Green function forward model according to a cylindrical geometric constraint, combining regularized inversion solving and radial basis interpolation reconstruction to obtain static electricity charge space distribution time series data; extracting multi-dimensional static electricity space-time risk features, performing threshold discrimination and weighted scoring to obtain a hazard level score, determining a static electricity hazard state according to a score interval, and generating a graded early warning information. The method can effectively eliminate detection interference, reconstruct a pipeline static electricity space-time distribution, realize static electricity risk fine grading early warning, and improve pipeline static electricity detection precision and safety prevention and control reliability.

Description

Technical Field

[0001] This invention belongs to the field of industrial electrostatic safety detection technology, and in particular relates to a method and system for detecting multi-point electrostatic distribution in pipelines. Background Technology

[0002] With the development of industrial electrostatic safety detection technology in industrial pipelines such as petrochemicals and energy transportation, pipeline electrostatic detection technology based on electrostatic potential has emerged. This technology collects electrostatic potential data by installing sensors on the inner wall of the pipeline, enabling online detection of the pipeline's electrostatic state. Traditional methods typically employ a single-point or small-scale sensor arrangement, collecting only local electrostatic potential values ​​and directly assessing electrostatic risk based on single-point potential thresholds.

[0003] The main processing flow of the traditional method is as follows: first, a sensor collects the local potential, then compares it with a preset safety threshold; once the threshold is exceeded, an alarm is triggered. This single-point sampling and threshold judgment approach is simple to implement and played a fundamental role in safety detection in early industrial scenarios.

[0004] However, current single-point threshold monitoring methods have significant limitations. First, the electrostatic charge on the inner wall of pipes is spatially non-uniformly distributed, and single-point acquisition cannot reflect the charge accumulation area and spatial distribution characteristics, easily missing high-risk areas far from the sensor. Second, traditional methods rely solely on static threshold judgments, failing to consider the spatiotemporal evolution of charge distribution and thus unable to capture dynamic hazard trends such as charge accumulation rate and increasing spatial gradient. Furthermore, coupling interference exists between sensor channels, and ambient temperature and humidity significantly affect potential measurement drift. Traditional methods lack effective decoupling and compensation mechanisms, resulting in detection accuracy and early warning reliability that are insufficient to meet the high safety requirements of industrial scenarios. Summary of the Invention

[0005] Based on this, it is necessary to provide a method and system for detecting electrostatic distribution at multiple points in pipelines. This method and system can obtain the time series of the spatial distribution of electrostatic charge on the inner wall of the pipeline through multi-point electrostatic potential sequence acquisition, channel decoupling, environmental adaptive compensation, Green's function inversion reconstruction based on cylindrical geometric constraints, and radial basis function interpolation. It can also extract spatiotemporal evolution features from the data for weighted hazard assessment and graded early warning, thereby comprehensively improving the detection accuracy and early warning reliability.

[0006] Firstly, this application provides a method for detecting electrostatic distribution at multiple points in a pipeline, including:

[0007] S1. Obtain the multi-point electrostatic potential data sequence synchronously collected by the electrostatic field sensor installed on the inner wall of the pipeline, the sampling time corresponding to each data point in the multi-point electrostatic potential data sequence, and the environmental parameter data sequence collected by the environmental parameter sensor in the pipeline during the multi-point electrostatic potential data sequence collection period;

[0008] S2. Based on the preset inter-channel coupling matrix, the multi-point electrostatic potential data sequence is decoupled to obtain the decoupled potential data sequence.

[0009] S3. Based on the environmental parameter data sequence, adaptive environmental compensation is performed on the decoupled potential data sequence to obtain the compensated potential data sequence.

[0010] S4. Discretize the inner wall of the pipe into multiple surface elements, and establish a forward model of the Green's function based on the cylindrical geometric constraints of the inner wall of the pipe. Combine the compensated potential vector at the current sampling time in the compensated potential data sequence to construct an inversion equation with the charge density of each surface element as the unknown. Solve the inversion equation using the Tikhonov regularization method to obtain the charge density of each surface element, and reconstruct the spatial distribution of electrostatic charge on the inner wall of the pipe through radial basis function spatial interpolation. Arrange the spatial distribution of electrostatic charge at each sampling time in chronological order to form a time series of electrostatic charge spatial distribution.

[0011] S5. Extract spatiotemporal evolution features from the time series of electrostatic charge spatial distribution, compare the spatiotemporal evolution features with preset danger thresholds, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic danger state based on the danger level score, and generate electrostatic early warning information based on the electrostatic danger state.

[0012] Secondly, this application also provides a pipeline multi-point electrostatic distribution detection system, comprising:

[0013] The data acquisition module is used to acquire multi-point electrostatic potential data sequences synchronously collected by electrostatic field sensors installed on the inner wall of the pipeline, the sampling time corresponding to each data point in the multi-point electrostatic potential data sequences, and environmental parameter data sequences collected by environmental parameter sensors in the pipeline during the multi-point electrostatic potential data sequence acquisition period.

[0014] The channel decoupling processing module is used to decouple the multi-point electrostatic potential data sequence according to the preset inter-channel coupling matrix to obtain the decoupled potential data sequence.

[0015] The environmental compensation processing module is used to perform adaptive environmental compensation on the decoupled potential data sequence based on the environmental parameter data sequence to obtain the compensated potential data sequence.

[0016] The electrostatic charge distribution reconstruction module discretizes the inner wall of the pipe into multiple surface elements and establishes a Green's function forward model based on the cylindrical geometric constraints of the inner wall of the pipe. Combining the compensated potential vector at the current sampling time in the compensated potential data sequence, an inversion equation is constructed with the charge density of each surface element as the unknown. The inversion equation is solved using the Tikhonov regularization method to obtain the charge density of each surface element. The spatial distribution of electrostatic charge on the inner wall of the pipe is reconstructed through radial basis function spatial interpolation. The spatial distribution of electrostatic charge at each sampling time is arranged in chronological order to form a time series of electrostatic charge spatial distribution.

[0017] The risk assessment and early warning module is used to extract spatiotemporal evolution characteristics from the time series of electrostatic charge spatial distribution, compare the spatiotemporal evolution characteristics with preset danger thresholds, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic hazard status based on the danger level score, and generate electrostatic early warning information based on the electrostatic hazard status.

[0018] The aforementioned method and system for detecting electrostatic distribution in pipelines eliminates coupling interference between electrostatic field sensors by combining multi-point electrostatic potential sequence acquisition with inter-channel coupling matrix decoupling. It utilizes environmental parameter data sequences for adaptive compensation, correcting for temperature and humidity drift at each time step, thus improving detection accuracy under complex operating conditions. Based on a Green's function forward model constrained by cylindrical geometry and Tikhonov regularized inversion, it reconstructs the complete spatial distribution of electrostatic charge on the inner wall of the pipeline, overcoming the limitation of single-point detection in reflecting charge accumulation areas. By constructing a time series of electrostatic charge spatial distribution, it extracts spatiotemporal evolution features such as maximum charge density, charge accumulation concentration, spatial gradient, and temporal change rate, and uses a weighted comprehensive scoring method to achieve graded early warning. This overcomes the limitations of traditional static threshold judgment, effectively capturing dynamic danger trends such as charge accumulation rate and increasing spatial gradient, significantly improving the foresight and reliability of electrostatic safety early warning. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram illustrating the application environment of a multi-point electrostatic distribution detection method for pipelines provided by the present invention.

[0021] Figure 2 A schematic flowchart of a multi-point electrostatic distribution detection method for pipelines provided by the present invention;

[0022] Figure 3 This is a schematic diagram of the structure of a multi-point electrostatic distribution detection system for pipelines provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0024] This application provides a method for detecting electrostatic distribution at multiple points in a pipeline, which can be applied to, for example... Figure 1 In the application environment shown, terminal 102 can communicate with server 101 via a network. A data storage system can store the data that server 101 needs to process. The data storage system can be integrated onto server 101 or placed in the cloud or on other network servers. Server 101 can receive multi-point electrostatic potential data sequences collected by multiple electrostatic field sensors deployed on the inner wall of the pipe, as well as environmental parameter data sequences collected by environmental parameter sensors. It performs processing such as channel decoupling, adaptive environmental compensation, forward modeling and Tikhonov regularized inversion reconstruction based on cylindrical geometric constraints, radial basis function spatial interpolation, and spatiotemporal evolution feature extraction and weighted hazard assessment. It generates electrostatic warning information and sends it to terminal 102 for display. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. Server 101 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0025] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting electrostatic distribution at multiple points in a pipeline is provided, which can be applied to... Figure 1 Taking server 101 as an example, the explanation includes the following steps:

[0026] S1. Obtain the multi-point electrostatic potential data sequence synchronously collected by the electrostatic field sensor installed on the inner wall of the pipeline, the sampling time corresponding to each data point in the multi-point electrostatic potential data sequence, and the environmental parameter data sequence collected by the environmental parameter sensor in the pipeline during the multi-point electrostatic potential data sequence collection period.

[0027] Specifically, the multi-point electrostatic potential data sequence refers to the time-series potential data set synchronously collected by multiple electrostatic field sensors deployed at different locations on the inner wall of the pipeline within a continuous sampling period. The multi-point electrostatic potential data sequence can comprehensively reflect the real-time distribution of electrostatic potential in different areas of the inner wall of the pipeline, accurately capturing the dynamic changes such as electrostatic accumulation and potential fluctuations generated during the flow of the medium inside the pipeline. The sampling time can refer to the precise timestamp corresponding to each set of electrostatic potential data. The environmental parameter data sequence can refer to the time-series data of the internal environment of the pipeline that is completely synchronized with the electrostatic potential acquisition period. The environmental parameter data sequence includes core environmental parameters such as temperature and humidity. Temperature and humidity are the core interference factors that affect the measurement accuracy of electrostatic field sensors and cause potential detection drift.

[0028] Optionally, server 101 can receive multi-point electrostatic potential data sequences and environmental parameter data sequences synchronously collected by multi-point electrostatic field sensors and environmental parameter sensors, and label the sensor number, collection point, collection time, equipment parameters, and pipeline operating condition information corresponding to each set of data. Server 101 can preprocess the collected multi-point electrostatic potential data sequences and environmental parameter data sequences, remove abnormal noise and abrupt bad values ​​caused by equipment vibration and electromagnetic interference during the collection process, interpolate and complete missing data, and unify the sampling frequency, data format, and timing accuracy of all sampled data to ensure that the potential data and environmental parameter data are completely matched in time dimension.

[0029] S2. Based on the preset inter-channel coupling matrix, the multi-point electrostatic potential data sequence is decoupled to obtain the decoupled potential data sequence.

[0030] Specifically, the inter-channel coupling matrix can refer to a fixed parameter matrix obtained through calibration experiments. The inter-channel coupling matrix can be used to characterize the electric field coupling interference relationship between multiple electrostatic field sensors on the inner wall of the pipeline, quantify the signal crosstalk and electric field superposition interference between different electrostatic field sensors, and is the core parameter for realizing multi-channel potential data separation and correction. The inter-channel coupling matrix can be obtained by applying gradient standard electrostatic potential excitation to each electrostatic field sensor individually under the standard calibration conditions of pipeline no load and no electrostatic interference, while the other electrostatic field sensors remain in no load state, and collecting the induced potential response data of the other acquisition channels one by one when each excitation channel is turned on. By fitting the potential coupling response ratio between each channel, the coupling coefficient between channels is calculated, and the inter-channel coupling matrix is ​​constructed by integrating all coupling coefficients.

[0031] Optionally, server 101 can acquire the excitation signal data applied for each excitation and the induced potential data synchronously collected by other electrostatic field sensors during multi-gradient standard electrostatic potential excitation calibration of all electrostatic field sensors under standard operating conditions of pipeline no-load, no medium, and no external electrostatic interference. Server 101 can calculate the coupling ratio coefficient between each electrostatic field sensor based on the excitation signal data of each excitation and the induced potential data of each electrostatic field sensor, integrate all coupling coefficients to generate a complete inter-channel coupling matrix, and repeatedly calibrate the matrix parameters in conjunction with the electrostatic field sensor layout spacing, pipeline diameter, and normal operating conditions, and then store the final calibrated matrix parameters in the system database. Server 101 can traverse the multi-point electrostatic potential data sequence at each sampling time and each data frame. For each sampling time, the multi-point electrostatic potential data at the current sampling time is organized into an original potential vector. By multiplying the inverse matrix of the inter-channel coupling matrix with the original potential vector, the decoupled potential data vector at the current sampling time is obtained. Server 101 can monitor the residuals and data fluctuations of each decoupling calculation in real time. It marks sampling frames where the decoupling residuals exceed a preset threshold or where data fluctuations are abnormal, retrieves the original data, and performs a second iteration of decoupling correction. The entire process records calibration parameters, decoupling operation parameters, abnormal correction records, and the final output results. Server 101 can arrange the decoupling potential data vectors from all sampling times in chronological order, integrate the clean potential data without channel crosstalk at all sampling times, and systematically assemble them into a standardized decoupling potential data sequence.

[0032] Preferably, the calculation formula for decoupling processing can be:

[0033]

[0034] in, This represents a multi-point electrostatic potential data vector at a given sampling time. This is the inter-channel coupling matrix. This is the inverse of the inter-channel coupling matrix. This is the decoupled potential data vector.

[0035] S3. Based on the environmental parameter data sequence, adaptive environmental compensation is performed on the decoupled potential data sequence to obtain the compensated potential data sequence.

[0036] Optionally, server 101 can pre-store the temperature sensitivity coefficient, humidity sensitivity coefficient, temperature-humidity coupling sensitivity coefficient, calibration reference temperature value, and calibration reference humidity value corresponding to each electrostatic field sensor. Server 101 can synchronously traverse the environmental parameter data sequence and decoupling potential data sequence at each sampling time, extract the real-time temperature value and real-time humidity value from the environmental parameter data sequence at the same sampling time as the current decoupling potential data, calculate the temperature deviation by subtracting the real-time temperature value from the calibration reference temperature value, and calculate the humidity deviation by subtracting the real-time humidity value from the calibration reference humidity value. Server 101 can multiply the temperature deviation by the temperature sensitivity coefficient to calculate the temperature drift compensation component; multiply the humidity deviation by the humidity sensitivity coefficient to calculate the humidity drift compensation component; multiply the temperature deviation by the humidity deviation to obtain the temperature-humidity coupling deviation; and then multiply the temperature-humidity coupling deviation by the temperature-humidity coupling sensitivity coefficient to calculate the temperature-humidity coupling drift compensation component. Server 101 can superimpose temperature drift compensation components, humidity drift compensation components, and temperature-humidity coupling drift compensation components to obtain the total environmental drift compensation amount at the current sampling time of each electrostatic field sensor. Server 101 can subtract the corresponding environmental drift compensation amount of each electrostatic field sensor from the decoupled potential data of each electrostatic field sensor to complete single data point correction. After traversing all sampling times and electrostatic field sensors and completing all data corrections, it integrates all corrected potential data according to the arrangement order and timing rules of the electrostatic field sensors, unifies and standardizes the data format, and finally generates a compensated potential data sequence.

[0037] S4. Discretize the inner wall of the pipe into multiple surface elements, and establish a forward model of the Green's function based on the cylindrical geometric constraints of the inner wall of the pipe. Combine the compensated potential vector at the current sampling time in the compensated potential data sequence to construct an inversion equation with the charge density of each surface element as the unknown. Solve the inversion equation using the Tikhonov regularization method to obtain the charge density of each surface element, and reconstruct the spatial distribution of electrostatic charge on the inner wall of the pipe through radial basis function spatial interpolation. Arrange the spatial distribution of electrostatic charge at each sampling time in chronological order to form a time series of electrostatic charge spatial distribution.

[0038] Optionally, server 101 can uniformly discretize the inner wall of the pipe into several regularly sized and arranged grid elements along the circumferential and axial directions based on the cylindrical geometric parameters of the inner wall, ensuring that the discretization accuracy meets the requirements for electrostatic distribution reconstruction. Server 101 can calculate the potential response value corresponding to the unit charge density of each grid element at each electrostatic field sensor placement location based on the cylindrical geometric constraints of the inner wall, and construct a Green's function forward model based on all potential response values. Server 101 can extract the compensated potential data sequence corresponding to all electrostatic field sensors at the current sampling time, organize it according to the fixed arrangement order of each electrostatic field sensor to construct a compensated potential vector, and combine it with the potential response matrix of the Green's function forward model to construct an inversion equation with the charge density of each discrete grid element as the unknown. Server 101 can use the Tikhonov regularization method to optimize the solution of the inversion equation, avoiding divergence problems in solving ill-conditioned equations by constraining the smoothness of the charge density distribution, and obtaining the charge density corresponding to each grid element at the current sampling time. Server 101 can perform radial basis function spatial interpolation on the charge density data of discrete surface elements based on the cylindrical geometric constraints of the inner wall of the pipe, using the geodesic distance of the curved surface as a distance metric, to fit a continuous and smooth electrostatic charge spatial distribution cloud map of the inner wall of the pipe. Server 101 repeats the above modeling, inversion, and interpolation reconstruction process for all sampling times one by one to obtain the electrostatic charge spatial distribution corresponding to each sampling time. The distribution is then arranged and integrated in chronological order of the sampling times to finally construct a complete time series of electrostatic charge spatial distribution.

[0039] S5. Extract spatiotemporal evolution features from the time series of electrostatic charge spatial distribution, compare the spatiotemporal evolution features with preset danger thresholds, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic danger state based on the danger level score, and generate electrostatic early warning information based on the electrostatic danger state.

[0040] Specifically, spatiotemporal evolution characteristics can refer to multidimensional core characteristic parameters that can quantify the degree of electrostatic risk. Spatiotemporal evolution characteristics include maximum charge density, charge concentration, global charge density spatial gradient, and global charge density temporal change rate.

[0041] Optionally, server 101 can pre-configure and store the independent hazard thresholds, feature weight coefficients, and hazard level scoring interval division standards corresponding to each spatiotemporal evolution feature quantity. Server 101 can traverse the time series of electrostatic charge spatial distribution sequentially, extracting the maximum charge density, charge concentration, global charge density spatial gradient, and global charge density temporal change rate at the current sampling moment frame by frame, completing the quantitative extraction of the entire set of spatiotemporal evolution feature quantities. Server 101 can compare each spatiotemporal evolution feature quantity with its corresponding preset hazard threshold one by one to determine whether each individual feature exceeds the standard, and generate a hazard state indication value based on the comparison result. Server 101 can call the preset first weight coefficient, second weight coefficient, third weight coefficient, and fourth weight coefficient to perform a weighted summation operation on each hazard state indication value to obtain the hazard level score corresponding to the current sampling moment. Server 101 can match and determine the hazard level score with the preset level interval to distinguish different electrostatic hazard states such as safe, caution, warning, and danger. Server 101 can generate electrostatic warning information corresponding to the level of electrostatic hazard based on the determined electrostatic hazard status. The warning information includes the risk location, risk level, risk evolution trend, and safety reminder content.

[0042] The aforementioned method for detecting electrostatic distribution in pipelines eliminates coupling interference between electrostatic field sensors by combining multi-point electrostatic potential sequence acquisition with inter-channel coupling matrix decoupling. It utilizes environmental parameter data sequences for adaptive compensation, correcting for temperature and humidity drift at each time step, thus improving detection accuracy under complex operating conditions. Based on a Green's function forward model constrained by cylindrical geometry and Tikhonov regularized inversion, it reconstructs the complete spatial distribution of electrostatic charge on the pipeline's inner wall, overcoming the limitation of single-point detection in reflecting charge accumulation areas. By constructing a time series of electrostatic charge spatial distribution, it extracts spatiotemporal evolution features such as maximum charge density, charge accumulation concentration, spatial gradient, and temporal rate of change, and employs a weighted comprehensive scoring method to achieve graded early warning. This method overcomes the limitations of traditional static threshold judgment, effectively capturing dynamic danger trends such as charge accumulation rate and increasing spatial gradient, significantly improving the foresight and reliability of electrostatic safety early warning.

[0043] In an exemplary embodiment, based on the environmental parameter data sequence, adaptive environmental compensation is performed on the decoupled potential data sequence to obtain the compensated potential data sequence, including the following steps:

[0044] S11. For each sampling time, extract the real-time temperature value and real-time humidity value from the environmental parameter data sequence. Divide the real-time temperature value by the calibration reference temperature value to obtain the temperature deviation. Divide the real-time humidity value by the calibration reference humidity value to obtain the humidity deviation.

[0045] Specifically, the calibration reference temperature value can refer to the standard ambient temperature value set according to the factory calibration specifications of the electrostatic field sensor and the normal operating temperature conditions of the pipeline. It is the standard reference temperature with the best temperature drift and detection accuracy. The calibration reference humidity value can refer to the standard ambient humidity value set according to the factory calibration specifications of the electrostatic field sensor and the normal operating dry conditions of the pipeline. It is the standard reference humidity with the best humidity drift and detection accuracy.

[0046] Optionally, the server can pre-store calibration reference temperature and humidity values, and retrieve the environmental parameter data sequence one by one, using each sampling time as a traversal unit. For each sampling time, the server can accurately extract the real-time temperature and humidity values ​​from the corresponding environmental parameter data sequence, ensuring that the extracted real-time environmental parameters precisely correspond to the sampling time without temporal misalignment. The server can perform difference calculations between the real-time temperature value corresponding to each sampling time and the preset calibration reference temperature value to obtain the temperature deviation for each sampling time. Similarly, the server can perform difference calculations between the real-time humidity value corresponding to each sampling time and the preset calibration reference humidity value to obtain the humidity deviation for each sampling time.

[0047] S12. Obtain the temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient of each electrostatic field sensor.

[0048] Specifically, the temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient can refer to quantitative characteristic parameters that characterize the changes in the detected potential of each electrostatic field sensor with temperature, humidity, and temperature-humidity coupling, respectively. They are the core coefficient parameters for calculating environmental drift compensation components and realizing adaptive environmental compensation of potential data. The temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient can refer to fixed parameters obtained by fitting multiple sets of gradient environmental calibration experiments, which can accurately quantify the potential detection drift law of each electrostatic field sensor under different environmental conditions.

[0049] Optionally, the server can pre-store the temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient corresponding to each electrostatic field sensor, which have been fitted, calibrated, and solidified through standard environmental calibration experiments. During environmental compensation calculations, the server can directly retrieve the temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient corresponding to each electrostatic field sensor.

[0050] S13. Multiply the temperature deviation by the temperature sensitivity coefficient to obtain the temperature drift compensation component. Multiply the humidity deviation by the humidity sensitivity coefficient to obtain the humidity drift compensation component. Multiply the temperature deviation by the humidity deviation to obtain the temperature-humidity coupling deviation. Multiply the temperature-humidity coupling deviation by the temperature-humidity coupling sensitivity coefficient to obtain the temperature-humidity coupling drift compensation component. Summate the temperature drift compensation component, humidity drift compensation component, and temperature-humidity coupling drift compensation component to obtain the environmental drift compensation amount of each electrostatic field sensor.

[0051] Optionally, the server can multiply the temperature deviation corresponding to the same sampling time and the same electrostatic field sensor by the temperature sensitivity coefficient to quantitatively obtain the temperature drift compensation component of the electrostatic field sensor. The server can also multiply the humidity deviation corresponding to the same sampling time and the same electrostatic field sensor by the humidity sensitivity coefficient to quantitatively obtain the humidity drift compensation component of the electrostatic field sensor. Furthermore, the server can multiply the temperature deviation and humidity deviation corresponding to the same sampling time and the same electrostatic field sensor to calculate the temperature-humidity coupling deviation, and then multiply this deviation by the temperature-humidity coupling sensitivity coefficient of the electrostatic field sensor to obtain the temperature-humidity coupling drift compensation component. The server can sum the temperature drift compensation component, humidity drift compensation component, and temperature-humidity coupling drift compensation component corresponding to the current sampling time of the current electrostatic field sensor to obtain the total environmental drift compensation amount for the corresponding sampling time. Finally, the server can iterate through all sampling times and perform component calculations and summations for all electrostatic field sensors to generate a dataset of environmental drift compensation amounts.

[0052] Preferably, the formula for calculating the environmental drift compensation amount can be:

[0053]

[0054] in, For the first An electrostatic field sensor in Environmental drift compensation at the sampling time, For temperature sensitivity coefficient, Humidity sensitivity coefficient This is the temperature and humidity coupling sensitivity coefficient. for Temperature deviation at any given time for Humidity deviation at any given time.

[0055] S14. Subtract the environmental drift compensation amount of each electrostatic field sensor from the decoupling potential value of each electrostatic field sensor in the decoupling potential data sequence to obtain the compensated potential value of each electrostatic field sensor. Arrange the compensated potential values ​​in the order of the electrostatic field sensors to form the compensated potential data sequence.

[0056] Specifically, the decoupling potential value can refer to the pure potential value of each electrostatic field sensor obtained after decoupling through the inter-channel coupling matrix, which eliminates electric field crosstalk and signal superposition interference between the channels of each electrostatic field sensor. This value is independent and free from channel coupling interference.

[0057] Optionally, the server can perform precise alignment and calibration calculations for each electrostatic field sensor at each sampling time. The server can retrieve the decoupling potential value and environmental drift compensation amount corresponding to the same sampling time and the same electrostatic field sensor, and subtract the environmental drift compensation amount matched to the electrostatic field sensor from the decoupling potential value of each electrostatic field sensor to obtain the compensated potential value of each electrostatic field sensor. After completing the potential calibration calculations for all sampling times and all electrostatic field sensors, the server can arrange and integrate all compensated potential values ​​in an orderly manner according to the preset deployment order of each electrostatic field sensor on the inner wall of the pipeline, unify the data storage format and time sequence arrangement rules, and finally form a compensated potential data sequence.

[0058] In an exemplary embodiment, the inner wall of the pipe is discretized into multiple surface elements, and a Green's function forward model is established based on the cylindrical geometric constraints of the inner wall of the pipe. Combining the compensated potential vector at the current sampling moment in the compensated potential data sequence, an inversion equation is constructed with the charge density of each surface element as the unknown. The Tikhonov regularization method is used to solve the inversion equation to obtain the charge density of each surface element, and the spatial distribution of electrostatic charge on the inner wall of the pipe is reconstructed through radial basis function spatial interpolation. The spatial distribution of electrostatic charge at each sampling moment is arranged in chronological order to form a time series of electrostatic charge spatial distribution, including the following steps:

[0059] S21. Discretize the inner wall of the pipe into grid-like surface elements along the circumferential and axial directions.

[0060] Optionally, the server can determine the number of circumferential and axial discrete parts based on the actual pipe diameter, axial length, and preset discretization accuracy threshold of the pipe's inner wall. The server can uniformly divide the inner wall of the pipe circumferentially and axially at uniform intervals, thus discretizing the complete and continuous cylindrical inner wall of the pipe into several rectangular grid-like elements.

[0061] S22. Based on the cylindrical geometric constraints of the inner wall of the pipe, calculate the potential response generated by the unit charge density of each surface element at the location of each electrostatic field sensor. Based on the potential response, establish a forward model of the Green's function.

[0062] Specifically, cylindrical geometric constraints refer to the fixed geometric boundary conditions brought about by the inherent cylindrical spatial structure of the inner wall of the pipe. Cylindrical geometric constraints include the relationship between fixed geometric parameters and spatial constraints such as constant pipe diameter, cylindrical axisymmetric characteristics, pipe wall surface curvature and axial regular extension structure. They are exclusive boundary constraints adapted to solving the electrostatic field of the pipe. Cylindrical geometric constraints can limit the distribution of electrostatic field, charge radiation path and potential spatial decay law.

[0063] Optionally, the server can retrieve cylindrical geometric parameters such as pipe diameter, axial length, and curvature of the pipe's inner wall, and combine them with the spatial coordinate information of each discrete surface element and the fixed layout coordinate information of each electrostatic field sensor to construct a spatial position mapping relationship based on cylindrical geometric constraints. The server can sequentially use each grid surface element as an independent excitation unit, set a unit charge density for each surface element, and calculate the potential response value generated by the unit charge density at the corresponding location of each electrostatic field sensor using the electrostatic field spatial potential solution formula. The server can traverse all discrete grid surface elements and all electrostatic field sensors to complete the potential response calculation at all points, integrate the corresponding potential response values ​​between all surface elements and electrostatic field sensors to construct a potential response matrix, and finally construct a Green's function forward model based on the spatial electric field mapping relationship fixed in this potential response matrix.

[0064] S23. Arrange the compensated potential values ​​of each electrostatic field sensor at the current sampling time in the compensated potential data sequence according to the order of the electrostatic field sensors, and construct the compensated potential vector; based on the Green's function forward model, construct the coefficient matrix with the potential response as the element, and construct the inversion equation with the charge density of each surface element as the unknown quantity with the compensated potential vector as the right-hand side.

[0065] Optionally, the server can lock onto the current single sampling moment to be solved and retrieve the compensated potential values ​​corresponding to all electrostatic field sensors at that sampling moment. The server can arrange all the compensated potential values ​​at the current sampling moment in an ordered manner according to the fixed deployment sequence of the electrostatic field sensors on the inner wall of the pipe, assembling a compensated potential vector with dimensions matching the number of electrostatic field sensors. The server can retrieve the already constructed Green's function forward model, extract all the discrete surface elements fixed within the Green's function forward model and the potential response values ​​corresponding to each electrostatic field sensor, and use the potential response of each surface element as matrix elements, arranging them in a regular manner according to the discrete order of the surface elements and the sensor arrangement order to construct a dimensionally matched and numerically complete inversion coefficient matrix. The server can set the compensated potential vector as the observation term on the right-hand side of the matrix equation, set the charge density of each discrete surface element as the unknown quantity in the equation, and combine the coefficient matrix to completely construct a standardized charge density inversion equation.

[0066] S24. Using the Tikhonov regularization method, the inversion equation solution is transformed into minimizing the objective function by constructing an objective function. By minimizing the objective function, the charge density of each surface element at the current sampling time is obtained.

[0067] Optionally, the server can optimize the constructed charge density inversion equation using the Tikhonov regularization method, combining data fitting terms and regularization constraint terms to construct a complete objective function. The server can use minimizing the residual between the compensated potential vector and the model-fitted potential vector as the data fitting objective, while introducing a regularization matrix to constrain the smoothness of the charge density distribution of each surface element, suppressing abrupt changes and oscillations in the numerical solution. The server can continuously iterate and update the iterative values ​​of the charge density of each surface element using a gradient iterative optimization algorithm, constantly reducing the function value of the objective function until the objective function converges to a preset minimum threshold, thus completing the minimization solution of the objective function. The server can use the converged optimal solution as the true charge density corresponding to each discrete surface element at the current sampling time.

[0068] S25. Based on the cylindrical geometric constraints of the inner wall of the pipe, the geodesic distance of the inner wall of the pipe is used as the distance metric of the radial basis function. The charge density of each surface element at the current sampling time is spatially interpolated by the radial basis function to obtain the spatial distribution of electrostatic charge on the inner wall of the pipe. The spatial distribution of electrostatic charge at each sampling time is arranged in chronological order to form a time series of electrostatic charge spatial distribution.

[0069] Optionally, the server can retrieve the geometric constraint parameters of the cylindrical inner wall of the pipe and the global spatial coordinate information to lock the charge density sample data of all discrete surface elements at the current sampling time. Based on the radial basis function interpolation model, the server can use each discrete surface element as the reference interpolation center point and the geodesic distance of the cylindrical inner wall surface as the distance metric of the radial basis function. Combining the pipe radius, the circumferential angle of the surface element, and the axial coordinates, the server calculates the shortest path distance between the point to be interpolated and the center point of each surface element. This distance is then substituted into the radial basis kernel function to fit and solve for the charge density interpolation values ​​of each spatial point on the global pipe wall, achieving continuous field reconstruction of the two-dimensional surface of the pipe inner wall and generating the spatial distribution of electrostatic charge on the pipe inner wall at the current sampling time. The server can iteratively execute the above interpolation reconstruction process for all sampling times to obtain the independent spatial distribution results of electrostatic charge corresponding to each sampling time in batches. The server can arrange and integrate the spatial distribution results of all frames of electrostatic charge according to the temporal order of the sampling times, unifying the spatiotemporal data format, and finally constructing a time series of electrostatic charge spatial distribution.

[0070] Preferably, the formula for calculating the radial basis function spatial interpolation can be:

[0071]

[0072] in, Any spatial position inside the pipe Interpolated charge density at the location, The total number of discrete surface elements. For the first The center coordinates of each element For radial basis kernel functions, Interpolation points on the cylindrical surface of the inner wall of the pipe With the Center point of each face element Geodesic distance between the curved surfaces For the first Undetermined weight coefficients of the basis functions It can be determined by solving a system of linear equations that satisfy the interpolation conditions at the centers of all facets.

[0073] Preferably, the formula for calculating the geodesic distance on a curved surface can be:

[0074]

[0075] in, The inner radius of the pipe. Points to be interpolated With the The minimum circumferential angle difference between the center points of each surface element Points to be interpolated With the The difference in axial coordinates between the center points of each element.

[0076] Preferably, The calculation formula can be:

[0077]

[0078] in, For the geodesic distance on the curved surface, This is the kernel width parameter.

[0079] In an exemplary embodiment, the objective function is calculated as follows:

[0080] ;

[0081] in, Let be the objective function. The coefficient matrix is ​​composed of the potential response. Let be the charge density vector composed of the charge densities of each surface element. For the compensated potential vector, The regularization matrix is ​​used to constrain the smoothness of the charge density distribution. This regularization matrix can be constructed using a second-order finite difference matrix. The square of the L2 norm, The regularization parameter can be determined using the L-curve method.

[0082] In an exemplary embodiment, spatiotemporal evolution features are extracted from the time series of the spatial distribution of electrostatic charge. These features are compared with a preset danger threshold, and a weighted comprehensive scoring method is used to calculate a danger level score. The electrostatic hazard status is determined based on the danger level score, and electrostatic early warning information is generated based on the electrostatic hazard status, including:

[0083] S31. For each sampling time, extract the absolute value of the charge density of each surface element at the current sampling time from the time series of electrostatic charge spatial distribution, and take the charge density with the largest absolute value as the maximum charge density; calculate the average charge density of the absolute values ​​of all surface elements, and compare the maximum charge density with the average charge density to obtain the charge concentration.

[0084] Optionally, the server can iterate through the time series of electrostatic charge spatial distribution at each sampling time, locking the charge density data of all discrete surface elements on the inner wall of the pipe corresponding to the current analysis time. The server can take the absolute value of the charge density value of each surface element to eliminate the interference of the positive or negative charge attribute on the determination of the degree of aggregation, and obtain the absolute value of the charge density corresponding to each surface element. The server can iterate and filter the absolute values ​​of the charge density of all surface elements at the current time, and extract the value with the largest absolute value of charge density as the maximum charge density at the current sampling time. The server can calculate the sum of the absolute values ​​of the charge density of all discrete surface elements at the current sampling time, and calculate the average charge density at the current time by combining it with the total number of surface elements. The server can calculate the charge aggregation concentration corresponding to the current sampling time by performing a ratio calculation between the maximum charge density and the average charge density.

[0085] S32. At the current sampling time, calculate the spatial difference of charge density between each group of adjacent surface elements, extract the geodesic distance of each group of adjacent surface elements, and take the ratio of the spatial difference of charge density to the geodesic distance as the spatial gradient of charge density; take the maximum value among the absolute values ​​of the spatial gradients of charge density of all adjacent surface elements as the global spatial gradient of charge density.

[0086] Optionally, the server can lock the spatial arrangement information and charge density data of all discrete surface elements on the inner wall of the pipe at the current sampling time. Based on the circumferential and axial adjacent topological relationships of the surface elements, it can match all adjacent surface element combinations one by one and generate a complete set of adjacent surface element groups. For each group of adjacent surface elements, the server can extract the charge density values ​​corresponding to the two surface elements respectively, and perform a difference operation on the charge density values ​​to obtain the spatial difference of charge density for each group. The server can retrieve the geometric parameters of the surface elements and extract the geodesic distance of the surface for each group of adjacent surface elements as the spatial reference spacing. The server can perform a ratio operation on the spatial difference of charge density and the geodesic distance corresponding to each group of adjacent surface elements to obtain the spatial gradient of charge density for each group of adjacent surface elements. The server can take the absolute value of the spatial gradient values ​​of charge density for all groups, iterate through and filter the maximum value among all absolute value data, and determine the maximum value as the global spatial gradient of charge density at the current sampling time.

[0087] S33. Subtract the charge density of the surface cell at the current sampling time from the charge density of the same surface cell at the previous sampling time to obtain the charge density time difference value. Calculate the time interval between the current sampling time and the previous sampling time. Take the ratio of the charge density time difference value to the time interval as the charge density time change rate. Take the maximum value among the absolute values ​​of the charge density time change rates of all surface cells as the global charge density time change rate.

[0088] Optionally, the server can perform temporal feature extraction operations on a sample-by-sample, discrete surface element basis, based on a time-series continuous spatial distribution of electrostatic charge. For the current sampling time, the server can accurately match the current charge density value of the corresponding surface element at the current sampling time with the charge density value of the same surface element at the previous adjacent sampling time, and perform a difference calculation to obtain the charge density time difference. The server can retrieve preset sampling time series parameters to calculate the standard time interval between the current sampling time and the previous sampling time. The server can perform a ratio calculation between the charge density time difference corresponding to each surface element and the time series time interval to obtain the charge density time change rate of all discrete surface elements. The server can take the absolute value of the charge density time change rate of all surface elements, iterate through and filter the largest value among all absolute value data, and determine the largest value as the global charge density time change rate at the current sampling time.

[0089] S34. When the maximum charge density exceeds the preset maximum charge density threshold, a maximum charge density danger state indication value of 1 is generated; otherwise, the value is 0. When the global charge density spatial gradient exceeds the preset spatial gradient threshold, a charge density spatial gradient danger state indication value of 1 is generated; otherwise, the value is 0. When the global charge density time change rate exceeds the preset time change rate threshold, a charge density time change rate danger state indication value of 1 is generated; otherwise, the value is 0. When the charge concentration exceeds the preset concentration threshold, a charge concentration danger state indication value of 1 is generated; otherwise, the value is 0.

[0090] Specifically, the preset maximum charge density threshold can refer to the critical value of the maximum allowable charge density at a single point on the pipe wall, obtained by fitting multiple sets of calibration experiments based on the pipeline's normal operating conditions, the electrostatic characteristics of the medium, and electrostatic discharge safety criteria. The preset maximum charge density threshold can be used to determine whether the local single-point charge accumulation in the pipeline exceeds the limit. The preset spatial gradient threshold can refer to the critical value of the maximum allowable gradient of charge density spatial change obtained by calibration based on the electrostatic distribution law of the pipeline's inner wall surface. The preset spatial gradient threshold can be used to determine whether there are abnormal spatial distribution risks such as sudden changes and abrupt changes in local charge on the pipe wall. The preset time change rate threshold can refer to the critical value of the maximum allowable rate of change of charge density over time obtained by calibration based on the dynamic accumulation and dissipation characteristics of pipeline electrostatics. The preset time change rate threshold can be used to determine whether there is a dynamic dangerous trend of rapid increase in electrostatic charge on the pipe wall. The preset aggregation concentration threshold can refer to the maximum allowable critical ratio of the degree of local charge aggregation obtained by calibration based on the uniform distribution benchmark of charge throughout the pipeline. The preset aggregation concentration threshold can be used to determine whether there are safety hazards of excessive local accumulation and severely uneven distribution of electrostatics in the pipeline.

[0091] Optionally, the server can pre-store four types of exclusive hazard judgment thresholds adapted to pipeline operating conditions: maximum charge density threshold, spatial gradient threshold, time change rate threshold, and aggregation concentration threshold. Each threshold is fixed and stored after being calibrated through pipeline electrostatic calibration experiments and fitting with operating condition safety thresholds. The server can compare the current maximum charge density with the preset maximum charge density threshold. If the maximum charge density exceeds the preset maximum charge density threshold, a maximum charge density hazard state indication value of 1 is generated; otherwise, a maximum charge density hazard state indication value of 0 is generated. The server can compare the current global charge density spatial gradient with the preset spatial gradient threshold. If the global charge density spatial gradient exceeds the preset spatial gradient threshold, a charge density spatial gradient hazard state indication value of 1 is generated; otherwise, a charge density spatial gradient hazard state indication value of 0 is generated. The server can compare the current global charge density time change rate with the preset time change rate threshold. If the global charge density time change rate exceeds the preset time change rate threshold, a charge density time change rate hazard state indication value of 1 is generated; otherwise, a charge density time change rate hazard state indication value of 0 is generated. The server can compare the current charge concentration with a preset concentration threshold. If the charge concentration exceeds the preset concentration threshold, a charge concentration danger status indication value of 1 is generated; otherwise, a charge concentration danger status indication value of 0 is generated.

[0092] S35. Multiply the maximum charge density hazard indicator value by a preset first weighting coefficient to obtain the first scoring component; multiply the charge density spatial gradient hazard indicator value by a preset second weighting coefficient to obtain the second scoring component; multiply the charge density time change rate hazard indicator value by a preset third weighting coefficient to obtain the third scoring component; multiply the charge aggregation concentration hazard indicator value by a preset fourth weighting coefficient to obtain the fourth scoring component; sum the first, second, third, and fourth scoring components to obtain the hazard level score.

[0093] Specifically, the first weighting coefficient can refer to the exclusive weighting coefficient of the maximum charge density risk dimension, which is calibrated based on the degree of harm and failure ratio of electrostatic breakdown and discharge accidents caused by single-point charge exceeding the limit in the pipeline. The first weighting coefficient can be used to quantify the hazard contribution weight of the risk of local peak charge exceeding the limit. The second weighting coefficient can refer to the exclusive weighting coefficient of the charge density spatial gradient risk dimension, which is calibrated based on the degree of harm and operating condition sensitivity of electrostatic anomalies caused by local charge spatial mutations and electric field distortions in the pipeline. The second weighting coefficient can be used to quantify the hazard contribution weight of the risk of charge spatial distribution mutations. The third weighting coefficient can refer to the exclusive weighting coefficient of the charge density time change rate risk dimension, which is calibrated based on the degree of evolution of sudden electrostatic hazards caused by rapid charge surges and dynamic instability in the pipeline. The third weighting coefficient can be used to quantify the hazard contribution weight of the risk of charge temporal dynamic evolution. The fourth weighting coefficient can refer to the exclusive weighting coefficient of the charge accumulation concentration risk dimension, which is calibrated based on the degree of harm of persistent electrostatic hazards caused by uneven charge distribution and local excessive accumulation in the pipeline. The fourth weighting coefficient can be used to quantify the hazard contribution weight of the risk of overall charge accumulation anomalies.

[0094] Optionally, the server can pre-fit, calibrate, and store mutually matching first, second, third, and fourth weighting coefficients through numerous pipeline electrostatic fault calibration experiments and operational risk hazard analysis. The server can retrieve the four types of hazard state indication values ​​corresponding to the current sampling time and perform weighted calculations for each. The server can multiply the maximum charge density hazard state indication value by the preset first weighting coefficient to calculate the first score component; multiply the charge density spatial gradient hazard state indication value by the preset second weighting coefficient to calculate the second score component; multiply the charge density time change rate hazard state indication value by the preset third weighting coefficient to calculate the third score component; and multiply the charge accumulation concentration hazard state indication value by the preset fourth weighting coefficient to calculate the fourth score component. The server can sum the calculated first, second, third, and fourth score components to obtain the electrostatic hazard level score corresponding to the current sampling time.

[0095] S36. Compare the hazard level score with the preset level range to obtain the electrostatic hazard status level of the pipeline, and generate electrostatic early warning information based on the electrostatic hazard status level.

[0096] Specifically, the preset level range can refer to a multi-level judgment threshold that is divided after combining pipeline normal operation safety standards, electrostatic fault critical conditions and equipment protection requirements, and through a large number of operating condition experiments and risk classification calibration. Each level of judgment threshold corresponds to a different degree of pipeline electrostatic hazard level, which is a standardized judgment basis for realizing the quantitative classification and differentiated early warning of electrostatic risks.

[0097] Optionally, the server can pre-store pre-defined multi-gradient level intervals. The boundary values ​​of each level interval are calibrated and solidified through pipeline electrostatic safety calibration experiments. The interval hierarchy is clearly defined, without overlap or gaps, adapting to the needs of electrostatic risk assessment under all pipeline operating conditions. The server can compare the hazard level score with the pre-defined level intervals. When the hazard level score equals zero, it is determined to be a safe state; when the hazard level score is greater than zero and less than or equal to the first pre-defined score threshold, it is determined to be a warning state; when the hazard level score is greater than the first pre-defined score threshold and less than or equal to the second pre-defined score threshold, it is determined to be a warning state; and when the hazard level score is greater than the second pre-defined score threshold, it is determined to be a dangerous state. This achieves full-coverage hierarchical control of electrostatic risk from safe, minor anomaly, moderate anomaly to severe danger, resulting in an electrostatic hazard status level. The server can call preset hierarchical warning rules to generate electrostatic warning information based on different electrostatic hazard status levels. The first and second pre-defined score thresholds can be set based on the maximum value and sum of the first, second, third, and fourth weighting coefficients.

[0098] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0099] Based on the same inventive concept, this application also provides a pipeline multi-point electrostatic distribution detection system for implementing the above-described pipeline multi-point electrostatic distribution detection method. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations of one or more pipeline multi-point electrostatic distribution detection system embodiments provided below can be found in the limitations of the pipeline multi-point electrostatic distribution detection method described above, and will not be repeated here.

[0100] In one exemplary embodiment, such as Figure 3 As shown, a pipeline multi-point electrostatic distribution detection system 300 is provided, comprising:

[0101] The data acquisition module 301 can be used to acquire multi-point electrostatic potential data sequences synchronously collected by electrostatic field sensors installed on the inner wall of the pipeline, the sampling time corresponding to each data point in the multi-point electrostatic potential data sequences, and environmental parameter data sequences collected by environmental parameter sensors in the pipeline during the multi-point electrostatic potential data sequence acquisition period.

[0102] The channel decoupling processing module 302 can be used to decouple a multi-point electrostatic potential data sequence according to a preset inter-channel coupling matrix to obtain a decoupled potential data sequence.

[0103] The environmental compensation processing module 303 can be used to perform adaptive environmental compensation on the decoupled potential data sequence based on the environmental parameter data sequence to obtain the compensated potential data sequence.

[0104] The electrostatic distribution reconstruction module 304 can be used to discretize the inner wall of the pipe into multiple surface elements, establish a Green's function forward model based on the cylindrical geometric constraints of the inner wall of the pipe, combine the compensated potential vector at the current sampling time in the compensated potential data sequence to construct an inversion equation with the charge density of each surface element as the unknown, solve the inversion equation using the Tikhonov regularization method to obtain the charge density of each surface element, and reconstruct the spatial distribution of electrostatic charge on the inner wall of the pipe through radial basis function spatial interpolation; arrange the spatial distribution of electrostatic charge at each sampling time in chronological order to form a time series of electrostatic charge spatial distribution.

[0105] The risk assessment and early warning module 305 can be used to extract spatiotemporal evolution characteristics from the time series of electrostatic charge spatial distribution, compare the spatiotemporal evolution characteristics with preset danger thresholds, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic danger state based on the danger level score, and generate electrostatic early warning information based on the electrostatic danger state.

[0106] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0107] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method of detecting multipoint static distribution in a pipeline, characterized by, The method includes: S1. Obtain the multi-point electrostatic potential data sequence synchronously collected by the electrostatic field sensor installed on the inner wall of the pipeline and the sampling time corresponding to each data point in the multi-point electrostatic potential data sequence, as well as the environmental parameter data sequence collected by the environmental parameter sensor in the pipeline during the multi-point electrostatic potential data sequence collection period; S2. According to the preset inter-channel coupling matrix, the multi-point electrostatic potential data sequence is decoupled to obtain a decoupled potential data sequence. S3. Based on the environmental parameter data sequence, perform adaptive environmental compensation on the decoupled potential data sequence to obtain the compensated potential data sequence; S4. Discretize the inner wall of the pipe into multiple surface elements, and establish a Green's function forward model based on the cylindrical geometric constraints of the inner wall of the pipe. Combine the compensated potential vector at the current sampling time in the compensated potential data sequence to construct an inversion equation with the charge density of each surface element as the unknown. Solve the inversion equation using the Tikhonov regularization method to obtain the charge density of each surface element, and reconstruct the electrostatic charge spatial distribution of the inner wall of the pipe through radial basis function spatial interpolation. Arrange the electrostatic charge spatial distribution at each sampling time in chronological order to form a time series of electrostatic charge spatial distribution. S5. Extract spatiotemporal evolution features from the time series of the spatial distribution of electrostatic charge, compare the spatiotemporal evolution features with a preset danger threshold, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic danger state based on the danger level score, and generate electrostatic early warning information based on the electrostatic danger state.

2. The method according to claim 1, characterized in that, The step of performing adaptive environmental compensation on the decoupled potential data sequence based on the environmental parameter data sequence to obtain a compensated potential data sequence includes: S11. For each of the sampling times, extract the real-time temperature value and the real-time humidity value from the environmental parameter data sequence, calculate the difference between the real-time temperature value and the calibration reference temperature value to obtain the temperature deviation, and calculate the difference between the real-time humidity value and the calibration reference humidity value to obtain the humidity deviation. S12. Obtain the temperature sensitivity coefficient, humidity sensitivity coefficient, and temperature-humidity coupling sensitivity coefficient of each electrostatic field sensor; S13. Multiply the temperature deviation by the temperature sensitivity coefficient to obtain a temperature drift compensation component; multiply the humidity deviation by the humidity sensitivity coefficient to obtain a humidity drift compensation component; multiply the temperature deviation by the humidity deviation to obtain a temperature-humidity coupling deviation; multiply the temperature-humidity coupling deviation by the temperature-humidity coupling sensitivity coefficient to obtain a temperature-humidity coupling drift compensation component; sum the temperature drift compensation component, the humidity drift compensation component, and the temperature-humidity coupling drift compensation component to obtain the environmental drift compensation amount of each electrostatic field sensor. S14. Subtract the environmental drift compensation amount of each electrostatic field sensor from the decoupling potential value of each electrostatic field sensor in the decoupling potential data sequence to obtain the compensated potential value of each electrostatic field sensor. Arrange the compensated potential values ​​in the order of the electrostatic field sensors to form the compensated potential data sequence.

3. The method according to claim 1, characterized in that, The inner wall of the pipe is discretized into multiple surface elements, and a Green's function forward model is established based on the cylindrical geometric constraints of the inner wall of the pipe. Combining the compensated potential vector at the current sampling time in the compensated potential data sequence, an inversion equation is constructed with the charge density of each surface element as the unknown. The inversion equation is solved using the Tikhonov regularization method to obtain the charge density of each surface element, and the spatial distribution of electrostatic charge on the inner wall of the pipe is reconstructed through radial basis function spatial interpolation. The spatial distribution of electrostatic charge at each sampling time is arranged in chronological order to form a time series of electrostatic charge spatial distribution, including: S21. Discretize the inner wall of the pipe into grid-like surface elements along the circumferential and axial directions; S22. Based on the cylindrical geometric constraints of the inner wall of the pipe, calculate the potential response generated by the unit charge density of each surface element at the location of each electrostatic field sensor, and establish the Green's function forward model based on the potential response. S23. Arrange the compensated potential values ​​of each electrostatic field sensor at the current sampling time in the compensated potential data sequence according to the order of the electrostatic field sensors to construct the compensated potential vector; based on the Green's function forward model, construct a coefficient matrix with the potential response as an element and the compensated potential vector as the right-hand side term to construct an inversion equation with the charge density of each surface element as the unknown. S24. Using the Tikhonov regularization method, by constructing an objective function, the inversion solution of the inversion equation is transformed into minimizing the objective function. By minimizing the objective function, the charge density of each surface element at the current sampling time is obtained. S25. Based on the cylindrical geometric constraints of the inner wall of the pipe, the geodesic distance of the inner wall of the pipe is used as the distance metric of the radial basis function. Radial basis function spatial interpolation is performed on the charge density of each surface element at the current sampling time to obtain the spatial distribution of electrostatic charge on the inner wall of the pipe. The spatial distribution of electrostatic charge at each sampling time is arranged in chronological order to form the time series of the spatial distribution of electrostatic charge.

4. The method according to claim 3, characterized in that, The formula for calculating the objective function is as follows: ； in, Let the objective function be... The coefficient matrix is ​​composed of the potential response. Let be the charge density vector formed by the charge densities of each of the aforementioned surface elements. The compensated potential vector, This is the regularization matrix used to constrain the smoothness of the charge density distribution. The square of the L2 norm, This is the regularization parameter.

5. The method according to claim 1, characterized in that, The process of extracting spatiotemporal evolution features from the time series of the spatial distribution of electrostatic charge, comparing the spatiotemporal evolution features with a preset danger threshold, calculating a danger level score using a weighted comprehensive scoring method, determining the electrostatic hazard status based on the danger level score, and generating electrostatic early warning information based on the electrostatic hazard status includes: S31. For each sampling time, extract the absolute value of the charge density of each surface element at the current sampling time from the electrostatic charge spatial distribution time series, and take the charge density with the largest absolute value of the charge density as the maximum charge density; calculate the average charge density of the absolute values ​​of the charge density of all surface elements, and compare the maximum charge density with the average charge density to obtain the charge concentration. S32. At the current sampling time, calculate the charge density spatial difference between each group of adjacent surface elements, extract the geodesic distance of each group of adjacent surface elements, and take the ratio of the charge density spatial difference to the geodesic distance as the charge density spatial gradient; take the maximum value among the absolute values ​​of the charge density spatial gradients of all adjacent surface elements as the global charge density spatial gradient. S33. Subtract the charge density of the surface element at the current sampling time from the charge density of the same surface element at the previous sampling time to obtain the charge density time difference value. Calculate the time interval between the current sampling time and the previous sampling time. Use the ratio of the charge density time difference value to the time interval as the charge density time change rate. Use the maximum value among the absolute values ​​of the charge density time change rates of all surface elements as the global charge density time change rate. S34. When the maximum charge density exceeds a preset maximum charge density threshold, a maximum charge density danger state indication value of 1 is generated; otherwise, the value is 0. When the global charge density spatial gradient exceeds a preset spatial gradient threshold, a charge density spatial gradient danger state indication value of 1 is generated; otherwise, the value is 0. When the global charge density time change rate exceeds a preset time change rate threshold, a charge density time change rate danger state indication value of 1 is generated; otherwise, the value is 0. When the charge concentration exceeds a preset concentration threshold, a charge concentration danger state indication value of 1 is generated; otherwise, the value is 0. S35. Multiply the maximum charge density hazard indicator value by a preset first weighting coefficient to obtain a first scoring component; multiply the charge density spatial gradient hazard indicator value by a preset second weighting coefficient to obtain a second scoring component; multiply the charge density time change rate hazard indicator value by a preset third weighting coefficient to obtain a third scoring component; multiply the charge aggregation concentration hazard indicator value by a preset fourth weighting coefficient to obtain a fourth scoring component; sum the first scoring component, the second scoring component, the third scoring component, and the fourth scoring component to obtain the hazard level score. S36. Compare the hazard level score with the preset level range to obtain the electrostatic hazard status level of the pipeline, and generate electrostatic warning information based on the electrostatic hazard status level.

6. A multi-point electrostatic distribution detection system for pipelines, characterized in that, The system includes: The data acquisition module is used to acquire multi-point electrostatic potential data sequences synchronously collected by electrostatic field sensors installed on the inner wall of the pipeline, the sampling time corresponding to each data point in the multi-point electrostatic potential data sequences, and environmental parameter data sequences collected by environmental parameter sensors in the pipeline during the acquisition period of the multi-point electrostatic potential data sequences. The channel decoupling processing module is used to decouple the multi-point electrostatic potential data sequence according to the preset inter-channel coupling matrix to obtain a decoupled potential data sequence. An environmental compensation processing module is used to perform adaptive environmental compensation on the decoupled potential data sequence based on the environmental parameter data sequence to obtain a compensated potential data sequence. The electrostatic distribution reconstruction module is used to discretize the inner wall of the pipe into multiple surface elements, establish a Green's function forward model based on the cylindrical geometric constraints of the inner wall of the pipe, combine the compensated potential vector at the current sampling time in the compensated potential data sequence to construct an inversion equation with the charge density of each surface element as the unknown, solve the inversion equation using the Tikhonov regularization method to obtain the charge density of each surface element, and reconstruct the electrostatic charge spatial distribution of the inner wall of the pipe through radial basis function spatial interpolation; arrange the electrostatic charge spatial distribution at each sampling time in chronological order to form an electrostatic charge spatial distribution time series. The risk assessment and early warning module is used to extract spatiotemporal evolution features from the time series of the spatial distribution of electrostatic charge, compare the spatiotemporal evolution features with a preset danger threshold, calculate the danger level score using a weighted comprehensive scoring method, determine the electrostatic danger state based on the danger level score, and generate electrostatic early warning information based on the electrostatic danger state.

Images