Three-dimensional transient magnetic field monitoring system based on sensor array attitude-position combined correction

Abstract

The invention discloses a three-dimensional transient magnetic field monitoring system based on sensor array attitude-position combined correction, and the system comprises magnetic field measurement assemblies which are composed of M * N * K three-dimensional sensor arrays and are distributed at equal intervals according to a preset space interval so as to collect transient magnetic field data in real time; the lower computer synchronously collects original data in parallel through the IIC bus and uploads the original data; the upper computer integrates an attitude error correction unit and a cooperative position compensation unit, eliminates the attitude error of the sensor through rotation matrix calculation based on a uniform magnetic field and a gradient magnetic field generated by a Helmholtz coil, and constructs a global optimization model to compensate position deviation. The data processing module adopts cubic spline interpolation to reconstruct continuous magnetic field distribution, and the display module realizes sub-layer visualization and time sequence playback of vector and scalar space-time distribution through a dynamic rendering technology; according to the system, high-precision monitoring of the transient magnetic field is achieved through a hardware parallel architecture and a joint correction algorithm, and the system is suitable for transient magnetic field dynamic analysis in the fields of power equipment, medical imaging and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic measurement, and particularly relates to a three-dimensional transient magnetic field monitoring system based on the joint correction of the attitude and position of a sensor array. Background Art

[0002] As an important part of modern science and technology, magnetic field monitoring technology is widely used in many fields, such as the electromagnetic environment monitoring of power systems, the magnetic field perception of aircraft in the aerospace field, the underground magnetic field detection in geological exploration, and the biomagnetic field measurement in biomedical engineering. With the continuous progress of technology, higher requirements have been put forward for the accuracy, efficiency, and data processing ability of magnetic field monitoring in various fields.

[0003] In the existing three-dimensional transient magnetic field monitoring technology, there are some problems and deficiencies. On the one hand, during the actual installation of the sensor array, due to the influence of various factors, such as the complexity of the installation environment and the error of the installation operation, there is a deviation between the actual position of the sensor and the theoretical position. On the other hand, there is inevitably an attitude error when the sensor is installed, that is, there is a rotational deviation between the measurement coordinate system of the sensor and the coordinate system of the actual magnetic field generating device. In addition, for the monitoring of transient magnetic fields, not only the spatial position and attitude information of the sensor need to be accurately obtained, but also real-time and high-precision data acquisition and processing of the rapidly changing magnetic field are required. In the data acquisition link of traditional monitoring systems, they are often limited by the data transmission rate and synchronous acquisition ability, and it is difficult to meet the requirements of the rapid change characteristics of transient magnetic fields. In terms of data processing, there is a lack of effective spatial interpolation and data reconstruction algorithms, and it is impossible to accurately recover the high-precision spatial magnetic field distribution from discrete sensor data, resulting in deficiencies in the spatio-temporal resolution of the monitoring results and being unable to clearly show the dynamic evolution process of transient magnetic fields. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide a three-dimensional transient magnetic field monitoring system based on the joint correction of the attitude and position of a sensor array, and solve the problems existing in the existing three-dimensional transient magnetic field monitoring system, such as insufficient installation accuracy of the sensor array, large attitude errors, and limited transient data acquisition and processing capabilities.

[0005] Technical solution: A three-dimensional transient magnetic field monitoring system based on joint attitude-position correction of a sensor array includes: a magnetic field measurement component composed of an M×N×K three-dimensional Hall effect sensor array, where the sensor array is equally spaced at a preset spatial interval for real-time acquisition of transient magnetic field data; a lower computer connected to the sensor array through an IIC bus, used to synchronously acquire the original magnetic field data of each sensor using a parallel data reading architecture and transmit the original data to the upper computer; an upper computer connected to the lower computer through a wired or wireless communication protocol, including: an attitude error correction unit that eliminates the sensor installation attitude error through rotation matrix calculation based on the uniform magnetic field generated by a Helmholtz coil; a collaborative position compensation unit that constructs a global optimization model based on the gradient magnetic field to compensate for the deviation between the theoretical position and the actual position of the sensor; a data processing module that uses a cubic spline interpolation algorithm to perform spatial reconstruction on the corrected discrete data to generate a continuous magnetic field distribution; a display module that displays the spatio-temporal distribution of magnetic field vectors / scalars in layers through dynamic rendering technology and supports time-series dynamic visualization.

[0006] Further, the attitude error correction unit includes:

[0007] A rotation matrix calculation module: used to calculate the true magnetic field vector B based on the orthogonal rotation matrix between the sensor measurement coordinate system S and the magnetic field generating device coordinate system G : Establish the rotation matrix between the sensor measurement coordinate system S and the magnetic field generating device coordinate system G th : The rotation matrix from coordinate system G to S Generate three orthogonal uniform magnetic fields B through the magnetic field generating device , measure the sensor output value B th , calculate the rotation matrix real Each element in: :

[0008] When applying a uniform magnetic field along the Z-axis , measure the sensor data Calculate:

[0009] r 11 = X1 / U, r 21 = Y1 / U, r 31 = Z1 / U

[0010] When applying a uniform magnetic field along the Y-axis , measure the sensor data Calculate:

[0011] r 12 = X2 / V, r 22 = Y2 / V, r 32 = Z2 / V

[0012] When a uniform magnetic field along the X-axis is applied the sensor data is measured Calculate:

[0013] r 13 = X3 / W, r 23 = Y3 / W, r 33 = Z3 / W

[0014] where U, V, and W are all arbitrary magnetic field values.

[0015] Transpose the matrix to obtain Perform a coordinate system transformation on the sensor measurement values and output the true magnetic field vector:

[0016]

[0017] The multi-sensor synchronization module calculates the position of the attitude matrix in real time for the magnetic field data of the sensor array through a parallel processing architecture: Through the parallel processing architecture, the rotation matrices of all sensors are synchronously solved to eliminate the overall attitude error of the array.

[0018] Furthermore, the collaborative position compensation unit includes: a deviation modeling module, based on the theoretical position of the sensor and the actual position of the residual function to construct a global optimization objective: Suppose there are n sensors in total, the theoretical position of the i-th sensor is the theoretical position the actual position is Construct the residual function:

[0019]

[0020] The global optimization module uses the least squares method to iteratively solve the optimal increment of the sensor position deviation and outputs the compensation parameters:

[0021] ΔP = -(J T J) -1 J T E

[0022] where

[0023]

[0024] When J T J is not invertible, by introducing a regularization parameter λ, the compensation parameter is corrected to:

[0025] ΔP = -(J T J + λI) -1 J T E

[0026] where λ is a small positive number and I is the identity matrix.

[0027] A real-time calibration module that adjusts the actual spatial coordinate parameters of the sensor array according to the compensation parameter Δp i

[0028] Furthermore, the data processing module is as follows: The discrete magnetic field data is divided into sub-intervals, and a cubic polynomial function S i (x, y, z) = a i (x - x i ) 3 + b i (x - x i ) 2 + c i (x - x i ) + d i ; where a i , b i , c i , d i are polynomial coefficients, and (x i , y i , z i ) are the node coordinates of the sub-interval; The polynomial coefficients are solved through interpolation conditions, derivative continuity constraints, and natural boundary conditions; The dynamic evolution data of the transient magnetic field is reconstructed by combining spatio-temporal correlation analysis.

[0029] The interpolation algorithm is implemented based on the following steps:

[0030] (1) The discrete magnetic field data points are divided into multiple sub-intervals according to spatial coordinates, and a cubic polynomial function is constructed within each sub-interval, in the form of:

[0031] S i (x, y, z) = a i (x - x i ) 3 + b i (x - x i ) 2 + c i (x - x i ) + d i

[0032] where a i , b i , c i , d i are polynomial coefficients, and (x i , y i , z i ) are the node coordinates of the sub-interval;

[0033] (2) Ensure that the function values at each node are consistent with the actual measured values through interpolation conditions:

[0034] S i (x i ,y i ,z i ) = B real (x i ,y i ,z i )

[0035] (3) Enforce the continuity of the first - order and second - order derivatives at the nodes of adjacent sub - intervals to ensure a smooth interpolation result;

[0036] (4) Adopt natural boundary conditions, that is, the second - order derivative at the global boundary is zero;

[0037] (5) Solve the linear equations through the three - moment algorithm to determine the polynomial coefficients of all sub - intervals;

[0038] Combine spatio - temporal correlation analysis to reconstruct the dynamic evolution data of the transient magnetic field.

[0039] Further, the sensor array is deployed in a three - dimensional orthogonal structure of M×N×K, where M, N, and K are positive integers, and each sensor is grouped and connected through the IIC bus, and each bus mounts sensors with a unique address.

[0040] Further, the display module is as follows: Based on the piecewise cubic spline interpolation and the reconstructed magnetic field data, generate the scalar spatio - temporal distribution of the transient magnetic field; Based on the corrected discrete magnetic field data, through dynamic rendering technology, allocate the total magnetic field vector and X / Y / Z components to independent layers and overlay and display them as a composite image to generate the vector spatio - temporal distribution map of the transient magnetic field; Integrate the time - axis control interface to support the time - series playback and analysis of the spatio - temporal distribution of the transient magnetic field.

[0041] Further, it also includes a data storage unit for storing the original data, correction parameters, and magnetic field distribution images, and supporting offline data playback.

[0042] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages: (1) By means of the attitude-position joint correction technology, the present invention eliminates the sensor installation error and significantly improves the magnetic field measurement accuracy. (2) Adopting a parallel data acquisition architecture, it supports parallel acquisition of multiple sensors and high-speed data processing, greatly improving the sampling rate and meeting the real-time monitoring requirements of transient magnetic fields. (3) Using the cubic spline interpolation algorithm and dynamic rendering technology, enables users to clearly observe the vector spatio-temporal distribution and scalar spatio-temporal distribution of transient magnetic fields, as well as the changes in the magnetic field at different time points. (4) The sensor array adopts an M×N×K three-dimensional structure, and the number and layout of sensors can be flexibly adjusted according to actual monitoring requirements, suitable for three-dimensional transient magnetic field monitoring scenarios of different scales and complexities, with wide applicability and good scalability. Description of the Drawings

[0043] Figure 1 is the block diagram of the three-dimensional transient magnetic field monitoring system of the present invention;

[0044] Figure 2 is the schematic diagram of the M×N×K three-dimensional Hall sensor array of the present invention;

[0045] Figure 3 is the physical diagram of the M×N×K three-dimensional Hall sensor array of the present invention;

[0046] Figure 4 is the physical diagram of the lower computer module of the present invention;

[0047] Figure 5 is the schematic diagram of the positional relationship between the sensor measurement coordinate system and the magnetic field generating device coordinate system of the present invention;

[0048] Figure 6 is the rotation matrix solution for attitude error correction and the calculation steps of magnetic field vector correction of the present invention;

[0049] Figure 7 is the flow chart of the global optimization algorithm for collaborative position compensation of the present invention;

[0050] Figure 8 is the data reconstruction flow chart of the piecewise cubic spline interpolation algorithm of the present invention;

[0051] Figure 9 is the dynamic rendering work flow chart of the display module of the present invention;

[0052] Figure 10 is the real-time three-dimensional magnetic field map of the dynamic rendering of the present invention;

[0053] Figure 11 is the real-time distribution map of the three-dimensional magnetic field layer by layer of the dynamic rendering of the present invention;

[0054] Figure 12This is the dynamic rendering effect diagram of the transient magnetic field vector / scalar sub - layer of the present invention. Specific embodiments

[0055] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0056] As Figure 1 shown, an embodiment of the present invention provides a three - dimensional transient magnetic field monitoring system based on the joint calibration of the attitude - position of a sensor array, including three parts: a three - dimensional sensor array, a lower computer, and an upper computer.

[0057] Figure 2 and Figure 3 In, the sensor array adopts a 4×4×4 three - dimensional orthogonal structure, with a total of 64 three - dimensional Hall effect sensors 1. The sensors are welded to a multi - layer PCB board 2 at an equal interval of 5 mm, and each layer of the PCB board corresponds to different heights of the Z - axis. Each sensor 1 communicates with the lower computer through the software IIC protocol, and its IIC address is configured as 4 independent addresses through hardware pins. To support parallel data acquisition, the system uses 16 groups of IIC buses, and each group of buses mounts 4 sensors (with unique addresses), constituting independent communication channels for the 4×4×4 array. The PCB board 2 is fixed by a non - magnetic bracket to ensure the spatial position accuracy of the array.

[0058] Figure 4 In, the lower computer uses a 32 - bit high - performance single - chip microcontroller with built - in software IIC driver. The parallel control of 16 groups of IIC buses is realized by simulating the IIC timing through GPIO. The software of the lower computer is configured to synchronously read the data of 16 groups of IIC buses; the original magnetic field data is encapsulated into data packets in the format of the sensor register (00H - 05H); the data packets are uploaded to the upper computer in real - time through the UART serial port (baud rate 230400).

[0059] The three - dimensional Helmholtz coil is composed of three pairs of orthogonal toroidal coils. The sensor array is positioned in the uniform magnetic field area at the center of the coil (error < 1 mm) for attitude error correction.

[0060] As Figure 5 shown is a schematic diagram of the position relationship between the sensor measurement coordinate system S and the magnetic field generating device coordinate system G caused by the installation attitude error.

[0061] As Figure 6 shown is a schematic diagram of the attitude error correction process, including the following steps:

[0062] (1) Align the sensor array coaxially with the Helmholtz coil, and apply uniform magnetic fields in the X / Y / Z - axis directions in sequence;

[0063] (2) Collect the output value B real of each sensor, and calculate the rotation matrix according to the following formula (Taking a single sensor as an example):

[0064] When a uniform magnetic field along the Z-axis is applied the sensor data is measured Calculate:

[0065] r 11 = X1 / U = 6.12 / 6.53 = 0.937

[0066] r 21 = Y1 / U = 0.32 / 6.53 = 0.049

[0067] r 31 = Z1 / U = -0.15 / 6.53 = -0.023

[0068] When a uniform magnetic field along the Y-axis is applied the sensor data is measured Calculate:

[0069] r 12 = X2 / V = 0.25 / 6.42 = 0.039

[0070] r 22 = Y2 / V = 6.08 / 6.42 = 0.947

[0071] r 32 = Z2 / V = 0.41 / 6.42 = 0.064

[0072] When a uniform magnetic field along the X-axis is applied the sensor data is measured Calculate:

[0073] r 13 = X3 / W = -0.18 / 6.71 = -0.027

[0074] r 23 = Y3 / W = 0.27 / 6.71 = 0.040

[0075] r 33 = Z3 / W = 6.45 / 6.71 = 0.961

[0076] Finally, the rotation matrix is obtained:

[0077]

[0078] (3) Convert the sensor measurement values to the device coordinate system through the transposed matrix :

[0079]

[0080] For example, the sensor measurement value is Converted to

[0081]

[0082] Such as Figure 7 The following is a schematic flow diagram of the collaborative position compensation of the present invention, including the following steps:

[0083] (1) Apply a gradient magnetic field through Helmholtz coils. Taking the same sensor as an example, the theoretical position is:

[0084] (2) Theoretical magnetic field The magnetic field measured by the sensor after attitude correction

[0085] (3) Construct the residual and Jacobian matrices:

[0086]

[0087] The Jacobian matrix J1 is composed of the partial derivatives of the magnetic field with respect to the position. For example:

[0088]

[0089] The same applies to the remaining components, and we get:

[0090]

[0091] (4) Solve for the position deviation increment:

[0092] Assume the global residual vector E = e1 and the Jacobian matrix J = J1. Through iterative calculation using the least squares method, we get:

[0093] Δp1 = (0.18mm, -0.22mm, 0.08mm) T

[0094] The actual position after compensation is updated to:

[0095] Generalize the above process to all sensors to complete the collaborative position compensation of the array

[0096] The upper computer is developed based on MATLAB. It receives the data packets from the lower computer through the serial port, parses them, and stores them as arrays, including the sensor positions, XYZ magnetic field components, and measurement times.

[0097] Such as Figure 8 As shown, for the corrected discrete data (4×4×4 array), the piecewise cubic spline interpolation algorithm is used to generate a continuous magnetic field distribution with a resolution of 0.5mm.

[0098] Such asFigure 9 The figure shows the working flow chart of the display module, and generates a spatio-temporal distribution image of the transient magnetic field intensity based on the interpolated data ( Figure 10 、 Figure 11 ). The three-dimensional vector direction is represented by a thick black arrow, and the X / Y / Z components are independently displayed by thin red / green / blue arrows respectively. The length of the arrow is linearly related to the magnetic field intensity, ensuring that the component arrows are spatially aligned with the total vector, and generating a spatio-temporal distribution image of the transient magnetic field vector. Dynamic update is achieved through fast graphics rendering technology, supporting fast forward / rewind control, as Figure 12 shown.

Claims

1. A three-dimensional transient magnetic field monitoring system based on joint correction of attitude and position of a sensor array, characterized in that, Comprising: A magnetic field measurement component, which is composed of an M×N×K three-dimensional Hall effect sensor array. The sensor array is equally spaced at a preset spatial interval and is used to collect transient magnetic field data in real time; A lower computer, which is connected to the sensor array through the IIC bus and is used to synchronously collect the original magnetic field data of each sensor by using a parallel data reading architecture and transmit the original data to the upper computer; An upper computer, which is connected to the lower computer through a wired or wireless communication protocol, including: an attitude error correction unit, which eliminates the sensor installation attitude error through rotation matrix calculation based on the uniform magnetic field generated by the Helmholtz coil; a collaborative position compensation unit, which constructs a global optimization model based on the gradient magnetic field to compensate for the deviation between the theoretical position and the actual position of the sensor; a data processing module, which uses a cubic spline interpolation algorithm to perform spatial reconstruction on the corrected discrete data to generate a continuous magnetic field distribution; a display module, which displays the spatio-temporal distribution of magnetic field vectors / scalars in layers through dynamic rendering technology and supports time series dynamic visualization.

2. The three-dimensional transient magnetic field monitoring system based on the combined attitude-position correction of a sensor array according to claim 1, wherein, The attitude error correction unit includes: a rotation matrix calculation module for calculating the true magnetic field vector B based on the orthogonal rotation matrix between the sensor measurement coordinate system S and the magnetic field generating device coordinate system G ; a multi-sensor synchronization module for performing real-time attitude matrix calculation on the magnetic field data of the sensor array through a parallel processing architecture. th ​ 3. A three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 2, characterized in that, The specific process of the rotation matrix calculation module is as follows: By using the preset magnetic field vector B in the coordinate system of the magnetic field generating device th and the sensor output value B real , the elements in the rotation matrix are solved and implemented through the following steps: First, apply a uniform magnetic field in the X / Y / Z axis directions Obtain the sensor output value Secondly, calculate the elements of the rotation matrix in it: r 11 = X1 / U, r 21 = Y1 / U, r 31 = Z1 / U r 12 = X2 / V, r 22 = Y2 / V, r 32 = Z2 / V r 13 = X3 / W, r 23 = Y3 / W, r 33 = Z3 / W Take the transpose of the rotation matrix as the calibration matrix, perform coordinate system transformation on the sensor measurement value, and output the true magnetic field vector: ​ 4. A three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 1, characterized in that, The collaborative position compensation unit includes: a position deviation modeling module that constructs a global optimization objective based on the residual function of the theoretical position of the sensor and the actual position ; a global optimization module that iteratively solves for the position deviation increment ΔP = -(J J + λI) T J -1 E; where J is a block diagonal matrix composed of Jacobian matrices, and λ is a regularization parameter; a real-time calibration module that dynamically adjusts the actual space coordinates of the sensor according to the compensation parameter Δp T ; i ​ 5. A three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 1, characterized in that The data processing module is as follows: divide the discrete magnetic field data into sub-intervals, and construct a cubic polynomial function S in each sub-interval i (x,y,z) = a i (x - x i ) 3 + b i (x - x i ) 2 + c i (x - x i ) + d i ; Among them, a i , b i , c i , d i are polynomial coefficients, and (x i , y i , z i ) are the coordinates of the sub-interval nodes; the polynomial coefficients are solved through interpolation conditions, derivative continuity constraints, and natural boundary conditions; combined with spatio-temporal correlation analysis, the dynamic evolution data of the transient magnetic field are reconstructed.

6. The three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 1, wherein The display module is specifically as follows: The total magnetic field vector and the X / Y / Z components are allocated to independent layers and displayed by dynamic rendering superposition; An integrated time axis control interface is supported for time series playback of spatio-temporal distribution.

7. A three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 1, characterized in that, The sensor array is deployed in an M×N×K three-dimensional orthogonal structure, where M, N, and K are positive integers, and each sensor is grouped and connected through the IIC bus, and each bus mounts sensors with a unique address.

8. A three-dimensional transient magnetic field monitoring system based on the combined correction of attitude and position of a sensor array according to claim 1, characterized in that, It also includes a data storage unit, which is used to store the original data, correction parameters, and magnetic field distribution images, and supports offline data playback.

Images