Three-dimensional control network adjustment method
Through the three-dimensional control network adjustment method, an independent coordinate system and error equation system are established in combination with multiple observations, which solves the problem of low elevation reliability of the control network during pumped storage power station construction, improves accuracy and reliability, and reduces external industries Work difficulty.
Patent Information
- Application Number
- CN202411090495.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the construction of pumped storage power stations, conventional total station or GNSS networking methods lead to low elevation reliability of the control network and poor network conditions, and the surveying and mapping instruments fail to effectively utilize the zenith distance and GNSS baseline vector, resulting in field operations Difficulty and increased workload.
The adjustment method of the three-dimensional control network is adopted. By establishing an independent coordinate system, combining horizontal angle, oblique distance, zenith distance, level height difference and GNSS baseline observation measurement, the calculation correction and adjustment processing are carried out, and the error equation system is established to solve the control point. and adjust the adjustment accuracy through observation weight.
It improves the elevation accuracy and reliability of the control network, reduces field workload and labor intensity, enhances the accuracy and reliability of the control network, and avoids the defect of choosing traffic difficult locations due to network type and communication requirements.
Smart Images

Figure CN119001783A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of control network adjustment, and in particular relates to a three-dimensional control network adjustment method. Background Art
[0002] At present, the construction control network of large hydropower stations is mostly built by total station conventional network layout or GNSS network layout method, and the elevation control adopts the precision leveling method, that is, the plane network and the elevation network are adjusted separately. Pumped storage power stations have the characteristics of small horizontal distance between the upper and lower reservoirs, large elevation difference, difficult line of sight between control network points, and difficulty in forming ideal graphics. Conventional methods usually solve the line of sight problem and graphic strength problem by adding transition points. Therefore, there are the following problems in the construction of the control network of pumped storage power stations: 1) Large elevation difference, poor network conditions, and single leveling line, resulting in low reliability of control network elevation; 2) In order to form an ideal network, some control network points are forced to be selected in difficult traffic locations, which brings about the problem of difficult field operations and doubled field workload. 3) The observations such as zenith distance and GNSS baseline vector that can be easily obtained under the existing surveying and mapping instruments and technical levels have not been effectively and comprehensively utilized in the adjustment calculation. Summary of the invention
[0003] In view of the defects in the prior art, the present invention provides a three-dimensional control network adjustment method, which can effectively solve the above problems.
[0004] The technical solution adopted by the present invention is as follows:
[0005] The present invention provides a three-dimensional control network adjustment method, comprising the following steps:
[0006] Step S1, establishing a three-dimensional independent coordinate system:
[0007] The center point is selected as the starting point of the control network in the control network, and the positive direction of the Z axis is the direction opposite to the vertical line direction of the starting point of the control network, which is the elevation direction; a certain normal high water level surface is selected as the projection surface of the plane coordinates, so as to establish a three-dimensional independent coordinate system;
[0008] Step S2, determining five observation quantities, namely: horizontal angle observation quantity, slant distance observation quantity, zenith distance observation quantity, level height difference observation quantity and GNSS baseline observation quantity;
[0009] For the control point i in the control network, relative to the sighting point k, one or more of the following five observation values are measured: horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik ;
[0010] Step S3, perform a correction process on each observation value to obtain the correction number of each observation value:
[0011] The horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik Perform the correction and get the horizontal angle correction value V Uik , slope distance correction number V Sik , Zenith distance correction V Bik , Level height difference correction factor V hik and GNSS baseline correction V Gik ;
[0012] Step S4, determine the approximate value of each observation:
[0013] Determine each horizontal angle observation value U separately ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The approximate value of the horizontal angle is obtained Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation
[0014] Step S5, determine the adjustment value of each observation:
[0015] Use formula (1) to formula (5) to determine each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The adjustment value of the horizontal angle is obtained Slope adjustment Zenith anomaly difference Level height difference adjustment value and GNSS baseline adjustment values
[0016]
[0017] Where: θ i is the azimuth of control point i, and are the approximate horizontal angles Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation number of corrections;
[0018] Step S6, establish the error equation:
[0019] Using formula (6) to formula (10), we can establish each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The error equation is:
[0020]
[0021] V hik =dz i +dz k -l hik (9)
[0022]
[0023] Where: ρ is the atmospheric density; and are the deviations of the control point i and the aiming point k in the x-direction, y-direction and z-direction respectively;
[0024] is the approximate value of horizontal distance; l Uik is the horizontal angle constant; l Sik is the slope distance constant term; l bik is the zenith distance constant term; l hik is the constant term of level height difference; l gik is the GNSS baseline constant term;
[0025] x i ,y i and z i , is the three-dimensional coordinate of the control point i to be solved;
[0026] x k ,y k and z k , is the three-dimensional coordinate of the aiming point k, which is a known value;
[0027] Step S7, multiple control points are arranged in the control network, and the number of control points whose three-dimensional coordinates are to be solved is m1, so the number of unknown parameters to be solved is m=3m1;
[0028] Count the number of observations of each type for all control points. Suppose a total of n1 horizontal angle observations U are obtained. ik , n2 slant range observations S ik , n3 zenith distance observations B ik , n4 level height difference observation values H ik , and n5 GNSS baseline observations G ik ; Therefore, the total number of observations is n = n1 + n2 + ... + n5;
[0029] The matrix representation of the error equation group established in step S6 is obtained as follows:
[0030]
[0031] Where: V is a correction matrix with n rows and 1 column; A is a coefficient matrix with n rows and m columns; E is a three-dimensional coordinate matrix (x, y, z) with m rows and 1 column; L is a constant term matrix with n rows and 1 column;
[0032] Step S8, solving the three-dimensional coordinates of each control point:
[0033] Step S8.1, the observation weights of the horizontal angle observation, the slant distance observation, the zenith distance observation, the level height difference observation and the GNSS baseline observation are respectively expressed as: P1, P2, P3, P4, P5; the initial values of the observation weights P1, P2, P3, P4, P5 are given;
[0034] Step S8.2, according to the matrix representation of the error equation group shown in formula (11), obtain the normal equation coefficient matrix N shown in formula (12) and the normal equation constant matrix W shown in formula (13):
[0035] The equation coefficient matrix N is:
[0036]
[0037] Wherein, the superscript T represents the transformation rank of the matrix; A1, A2, A3, A4, A5 represent the coefficient matrix components of the coefficient matrix A regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0038] N1, N2, N3, N4, N5 represent the coefficient matrix of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0039] The equation constant matrix W is:
[0040]
[0041] Where: l1, l2, l3, l4, l5 represent the constant term components of the constant term matrix L regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0042] W1, W2, W3, W4, and W5 represent the constant arrays of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation, and GNSS baseline observation, respectively;
[0043] Step S8.3, the solution vector of the unknown parameters E = (X, Y, Z) is: Where: Q XX is the cofactor matrix;
[0044] Thus, the three-dimensional coordinates of each control point are obtained;
[0045] Step S8.4, using formula (14), obtain the unit weight variance of the adjustment value
[0046]
[0047] Among them, P = [P1, P2, P3, P4, P5] is the observation weight matrix; V = [V Uik ,V Sik ,V Bik ,V hik ,V Gik ], is the observation correction matrix; r is the total number of redundant observations;
[0048] Step S8.5, using formula (15), calculate the posterior variance of each observation j Among them, j = 1, 2, 3, 4, 5, representing the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0049]
[0050] Where: P j represents the observation weight of observation j; V j represents the correction number of observation j; R j is the number of redundant observations of observation j;
[0051] Step S8.6, determine the posterior variance of each observation j Is the deviation from the prior variance of each observation j within the set range? If so, end the process and output the three-dimensional coordinates of each control point solved by step S8.3; otherwise, adjust the observation weights P1, P2, P3, P4, P5 of each observation and return to step S8.2 to loop.
[0052] Preferably, in step S3, the horizontal angle correction value V Uik 0; GNSS baseline correction number V Gik is 0;
[0053] For the slope distance observation value S ik , respectively make corrections for ruler length, meteorological correction, and channel curvature, and then return to the mark center;
[0054] For the zenith distance observation value B ik , respectively, the zenith distance observation value B ik Correction normalized to the center of the mark and spherical aberration correction;
[0055] For the level height difference observation value H ik , if H ik If the distance is greater than 500m, corrections are made for the length of the rod, non-parallelism of the level surface and gravity anomaly.
[0056] Preferably, in step S4, the following method is used to obtain an approximate value of each observation value:
[0057] The control network points for which approximate values need to be obtained are jointly measured with the plane coordinates and elevations of the known points. Each known point is used as a joint measurement position. Therefore, there are no less than three joint measurement positions, and they are evenly distributed in the entire survey area. Then a similarity transformation is performed to make the sum of the squares of the residuals of the coordinate components of each point extremely small, so as to obtain various approximate values of the control network points.
[0058] The three-dimensional control network adjustment method provided by the present invention has the following advantages:
[0059] 1) Geometric leveling is used to control the elevation accuracy of the entire network, while redundant observations of trigonometric leveling are used to enhance the reliability of elevation control; 2) The location of control network points can be selected more freely, and it is not necessary to select locations with difficult transportation and inconvenient use simply because of network type and line of sight requirements, thereby reducing field workload and labor intensity; 3) Five types of observations, namely horizontal angle observations, slant distance observations, zenith distance observations, level height difference observations and GNSS baseline observations, can be used to carry out control network adjustment, improve the accuracy of the control network and enhance reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 A schematic flow chart of a three-dimensional control network adjustment method provided by the present invention;
[0061] Figure 2 A coordinate system establishment diagram for a three-dimensional control network adjustment method provided by the present invention. DETAILED DESCRIPTION
[0062] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0063] The invention relates to the technical field of control network adjustment, and in particular to a hybrid three-dimensional network adjustment method combining two-dimensional plane and one-dimensional elevation, which is applied to small-scale construction / monitoring control network measurement in pumped storage power stations or hydropower engineering hubs.
[0064] See also Figure 1 The present invention provides a three-dimensional control network adjustment method, comprising the following steps:
[0065] Step S1, establishing a three-dimensional independent coordinate system:
[0066] See also Figure 2 , select the center point in the control network as the starting point of the control network, and take the positive direction of the Z axis in the direction opposite to the vertical line direction of the starting point of the control network as the elevation direction; take a selected normal high water level surface as the projection surface of the plane coordinates, so as to establish a three-dimensional independent coordinate system;
[0067] Specifically, in a specific small range, a plane is used instead of a sphere to find out the deformation of its length and angle. When it is agreed that the coverage of the construction control network of a pumped storage power station does not exceed 20 square kilometers, the distance from the edge to the center of the network is about 2.5km. Through calculation, it can be known that the error of the arc length (or chord length) of 2.5km on the sphere projected onto the tangent plane is about 0.1mm; in addition, the spherical angle calculated by an equilateral triangle with a side length of 2.5km is 0.014", so the error brought to each angle is less than 0.005".
[0068] Therefore, the three-dimensional coordinate system of the present invention is to select a point located at or near the center of the control network as the starting point of the control network, and take the direction opposite to the vertical direction of this point as the positive direction of the Z axis, that is, the elevation direction; take the selected normal high water level surface as the projection surface of the plane coordinates, and directly calculate the plane coordinates from the side length and direction on the spherical surface; the water level surface is approximately replaced by the spherical surface, and the height difference between the two points is still based on the spherical surface. The three-dimensional coordinate system established in this way is a mixed coordinate system of the plane coordinate system and the elevation system. The definition of the X, Y, and Z axes is referenced in the attached Figure 1 .
[0069] Step S2, determining five observation quantities, namely: horizontal angle observation quantity, slant distance observation quantity, zenith distance observation quantity, level height difference observation quantity and GNSS baseline observation quantity;
[0070] For the control point i in the control network, relative to the sighting point k, use the measuring instrument to collect and measure the observation values of the following five observation quantities, one or more of which are obtained: horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik ;
[0071] Step S3, perform a correction process on each observation value to obtain the correction number of each observation value:
[0072] The horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik Perform the correction and get the horizontal angle correction value V Uik , slope distance correction number V Sik , Zenith distance correction V Bik , Level height difference correction factor V hik and GNSS baseline correction V Gik ;
[0073] The calculation method for each observation value can be:
[0074] (a) Horizontal angle observation value U ik There is no correction for the alignment and station centering. In addition, since the calculation is made to the spherical surface closest to the quasi-geoid, and the slight fluctuations of the level surface are ignored, the vertical line and the normal line are consistent, there is no vertical line deviation, and there is no correction for the vertical line deviation. The horizontal angle observation value U measured by the station is ik It can directly participate in three-dimensional adjustment without any modification.
[0075] (b) Slope range observation value S ik The calculation is as follows: make ruler length correction, meteorological correction, channel curvature correction respectively, and calculate to the mark center.
[0076] (c) Zenith distance observation value B ik Calculation: The zenith distance observation value B ik Correction normalized to the center of the mark and spherical aberration correction.
[0077] (d) Level height correction factor V hik Imputation
[0078] For pumped-storage power stations with smaller level height differences, it is generally required to conduct third-order geometric leveling without adding correction for non-parallelism of the level surface; when the height difference is greater than 500m, precise geometric leveling is generally adopted, and therefore correction for length of the rod, non-parallelism of the level surface and gravity anomaly are added.
[0079] (e) GNSS baseline correction V Gik Imputation and Correction
[0080] After baseline solution, the GNSS observation values have been converted into chord lengths between the centers of landmarks, just like conventional side measurement, and can be directly used in three-dimensional adjustment, while paying attention to the scale ratio correction with the conventional side.
[0081] Step S4, determine the approximate value of each observation:
[0082] Determine each horizontal angle observation value U separately ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The approximate value of the horizontal angle is obtained Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation
[0083] In step S4, the following method is used to obtain the approximate value of each observation:
[0084] The control network points for which approximate values need to be obtained are jointly measured with the plane coordinates and elevations of the known points. Each known point is used as a joint measurement position. Therefore, there are no less than three joint measurement positions, and they are evenly distributed in the entire survey area. Then a similarity transformation is performed to make the sum of the squares of the residuals of the coordinate components of each point extremely small, so as to obtain various approximate values of the control network points.
[0085] Step S5, determine the adjustment value of each observation:
[0086] Use formula (1) to formula (5) to determine each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The adjustment value of the horizontal angle is obtained Slope adjustment Zenith anomaly difference Level height difference adjustment value and GNSS baseline adjustment values
[0087]
[0088] Where: θ i is the azimuth of control point i, and are the approximate horizontal angles Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation number of corrections;
[0089] Step S6, establish the error equation:
[0090] Using formula (6) to formula (10), we can establish each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The error equation is:
[0091]
[0092]
[0093] V hik =dz i +dz k -l hik (9)
[0094]
[0095] Where: ρ is the atmospheric density; and are the deviations of the control point i and the aiming point k in the x-direction, y-direction and z-direction respectively;
[0096] is the approximate value of horizontal distance; l Uik is the horizontal angle constant; l Sik is the slope distance constant term; l bik is the zenith distance constant term; l hik is the constant term of level height difference; l gik is the GNSS baseline constant term;
[0097] x i ,y i and z i , is the three-dimensional coordinate of the control point i to be solved;
[0098] x k ,y k and z k , is the three-dimensional coordinate of the aiming point k, which is a known value;
[0099] Step S7, multiple control points are arranged in the control network, and the number of control points whose three-dimensional coordinates are to be solved is m1, so the number of unknown parameters to be solved is m=3m1;
[0100] Count the number of observations of each type for all control points. Suppose a total of n1 horizontal angle observations U are obtained. ik , n2 slant range observations S ik , n3 zenith distance observations B ik , n4 level height difference observation values H ik , and n5 GNSS baseline observations G ik ; Therefore, the total number of observations is n = n1 + n2 + ... + n5;
[0101] The matrix representation of the error equation group established in step S6 is obtained as follows:
[0102]
[0103] Where: V is a correction matrix with n rows and 1 column; A is a coefficient matrix with n rows and m columns; E is a three-dimensional coordinate matrix (x, y, z) with m rows and 1 column; L is a constant term matrix with n rows and 1 column;
[0104] Step S8, solving the three-dimensional coordinates of each control point:
[0105] Step S8.1, the observation weights of the horizontal angle observation, the slant distance observation, the zenith distance observation, the level height difference observation and the GNSS baseline observation are respectively expressed as: P1, P2, P3, P4, P5; the initial values of the observation weights P1, P2, P3, P4, P5 are given;
[0106] Step S8.2, according to the matrix representation of the error equation group shown in formula (11), obtain the normal equation coefficient matrix N shown in formula (12) and the normal equation constant matrix W shown in formula (13):
[0107] The equation coefficient matrix N is:
[0108]
[0109] Wherein, the superscript T represents the transformation rank of the matrix; A1, A2, A3, A4, A5 represent the coefficient matrix components of the coefficient matrix A regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0110] N1, N2, N3, N4, N5 represent the coefficient matrix of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0111] The equation constant matrix W is:
[0112]
[0113] Where: l1, l2, l3, l4, l5 represent the constant term components of the constant term matrix L regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0114] W1, W2, W3, W4, and W5 represent the constant arrays of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation, and GNSS baseline observation, respectively;
[0115] Step S8.3, the solution vector of the unknown parameters E = (X, Y, Z) is: Where: Q XX is the cofactor matrix;
[0116] Thus, the three-dimensional coordinates of each control point are obtained;
[0117] Step S8.4, using formula (14), obtain the unit weight variance of the adjustment value
[0118]
[0119] Among them, P = [P1, P2, P3, P4, P5] is the observation weight matrix; V = [V Uik ,V Sik ,V Bik ,V hik ,V Gik ], is the observation correction matrix; r is the total number of redundant observations;
[0120] Step S8.5, using formula (15), calculate the posterior variance of each observation j Among them, j = 1, 2, 3, 4, 5, representing the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively;
[0121]
[0122] Where: P j represents the observation weight of observation j; V j represents the correction number of observation j; R j is the number of redundant observations of observation j;
[0123] Step S8.6, determine the posterior variance of each observation j Is the deviation from the prior variance of each observation j within the set range? If so, end the process and output the three-dimensional coordinates of each control point solved by step S8.3; otherwise, adjust the observation weights P1, P2, P3, P4, P5 of each observation and return to step S8.2 to loop.
[0124] Specifically, if the observation equation remains unchanged and different prior variances are given, different adjustment values and cofactor matrices will be obtained. Only when the prior variance and the posterior variance are consistent can it be considered that the accuracy requirements are met. When correcting the measurement weights, the initial variance can be used to determine the weights for adjustment, and then the variance component estimates can be obtained using the Helment method. The weights are modified until the variances of the various observations become equal.
[0125] By using the three-dimensional adjustment data processing method provided by the present invention, the horizontal angle, slant distance, zenith distance, level height difference and GNSS baseline vector five types of observations are adjusted directly instead of the function of the observation quantity. The solution method is more reasonable and ensures that the observation accuracy is not lost. Compared with the traditional plane and elevation separate adjustment method, the present invention has the following advantages in the actual construction of the construction control network of the pumped storage power station:
[0126] 1) When calculating the control network adjustment, trigonometric height, GNSS baseline and other observations are added. Compared with the single leveling method, it can more effectively discover the possible systematic errors in the height measurement of large height difference terrain, and greatly improve the reliability of height control measurement;
[0127] 2) Construction control points can be arranged according to project requirements without over-emphasizing the type of control network. Point selection is more convenient, thereby reducing the difficulty of burying and observing some control points, reducing workload, and reducing the difficulty of field data collection;
[0128] 3) Observations such as trigonometric height and GNSS baseline enrich the field data collection methods of the engineering control network, and their data can be effectively used during adjustment. Therefore, the three-dimensional network layout method is a better choice for establishing the construction control network of the energy storage power station, which not only avoids some technical indicators of the conventional method exceeding the requirements of the specification, but also improves the accuracy and reliability of the final results. This is the technical advantage of three-dimensional control network adjustment in pumped storage power station projects.
[0129] Experimental research and applied statistics were conducted based on four projects, namely, Zhanghewan Pumped Storage Power Station in Hebei, Langyashan Pumped Storage Power Station in Anhui, Xilongchi Pumped Storage Power Station in Shanxi, and Hohhot Pumped Storage Power Station in Inner Mongolia. The construction control networks of Zhanghewan Pumped Storage Power Station in Hebei and Langyashan Pumped Storage Power Station in Anhui were laid out using traditional two-dimensional plane and one-dimensional elevation networks, which were verified using three-dimensional networks. The construction control networks of Xilongchi Pumped Storage Power Station in Shanxi and Hohhot Pumped Storage Power Station in Inner Mongolia were laid out using the three-dimensional control network adjustment method provided by the present invention, which were verified using two-dimensional plane networks and one-dimensional elevation networks. In addition to verifying the influence of plane coordinate results and point accuracy, the accuracy and reliability of elevation control were also verified.
[0130] The statistics of the four projects are as follows:
[0131]
[0132]
[0133] It can be seen from the above table that when the observation is carried out without taking any special measures, the requirements can be met after the three-dimensional control network adjustment processing according to the present invention:
[0134] (1) The unit weight variance meets the accuracy index specified in the design. Only the mean error of the horizontal angle of the construction control network of Xilongchi, Shanxi Province is slightly larger. However, the unit weight mean error after adjustment should be less than 1.5 times of the unit weight observation mean error. Therefore, it still meets the requirements.
[0135] (2) The mean error of the weakest point is within the design accuracy range, including the error of the plane coordinate component, the point error, and the elevation error. Compared with the plane network and the elevation network, the average point accuracy and the weakest point accuracy have been improved. The plane position accuracy has increased by an average of 18%, and the elevation accuracy has increased by more than 70%. The results are statistically shown in the following table:
[0136] Statistics of the four project point accuracy indicators
[0137]
[0138] (3) There are certain differences in the coordinates of the two adjustment results, but both are within the range of error of the specified point positions.
[0139] (4) Among the four projects, the first three power stations have been completed and are generating electricity. The civil construction of the Hohhot Pumped Storage Power Station in Inner Mongolia has been preliminarily completed, and the underground caverns have been properly connected, indicating that the control results have fully met the project requirements.
[0140] The statistics of the design error, adjustment error, redundant observation components and corresponding weights of various observation quantities are as follows:
[0141]
[0142] The variance consistency test is used to determine whether to add the vertical deviation parameter. If the unit weight variance μ 2 The mathematical expectation is equal to the prior variance σ 2 , it means that there is no need to introduce the vertical deviation parameter, otherwise the vertical deviation parameter should be introduced. Since the present invention adopts an advanced variance component estimation weighting method, the posterior variance is consistent with the prior variance, so it is only necessary to check the unit weighted mean error after the adjustment of various observations. If the test result meets the design requirements, it can be considered that the adjustment value does not have the influence of gross errors, so it can be considered that there is no need to introduce the vertical deviation parameter. At the same time, due to the large number of redundant observations, it is easy to find observations with gross errors, and eliminating these gross errors can improve the adjustment quality of the network. However, it is not easy to distinguish whether the cause of the gross error is caused by the measurement itself or by the vertical deviation or vertical refraction difference.
[0143] Due to different natural conditions, the number of gross errors in the four projects is quite different. Usually, the mean error greater than twice the adjustment value is treated as a gross error. Zhanghewan has 2 gross errors in the zenith distance; Xilongchi has 1 gross error in the horizontal angle, 6 gross errors in the slant distance, and 4 gross errors in the zenith distance. The gross error elimination rate is about 5%, of which the elimination rate of the zenith distance is 6%; Langya Mountain and Hohhot have no gross errors. The unit weighted variances of various observation quantities of the four projects are better than the design specifications except for the horizontal angle of Xilongchi, indicating that the processing results of the three-dimensional adjustment data are normal.
[0144] The key to 3D data processing is the reduction (rough calculation) and adjustment of various observation quantities, among which the most noteworthy is the zenith distance. Refraction error affects both horizontal angle and zenith distance, but lateral refraction is generally easier to avoid, while vertical refraction is not easy to avoid. This is the main reason why the observation accuracy of horizontal angle is usually higher than that of zenith distance. In addition, there are other effects such as vertical deviation. If the mean error of zenith distance after adjustment meets the design requirements and there are few gross errors in the residual, then it can be said that its rough calculation and adjustment processing are reasonable. In the construction control networks of the four projects, the mean error of zenith distance adjustment is better than the design accuracy requirements, and the elimination rate of gross errors is also very low, which shows that the simplified processing method for vertical refraction error and vertical deviation is feasible and successful.
[0145] In addition, it can be seen from the actual accuracy statistics of the four projects that although the difficulty of the Xilongchi project is comparable to that of the Hohhot project, and the accuracy of the final results can well meet the requirements of the project, the former is inferior to the latter. In order to do a good job in establishing the construction control network of Fengning and other pumped storage power stations in the future, the inventor has conducted a systematic analysis of the causes of its occurrence and believes that the following three aspects of experience and lessons are worth learning:
[0146] First, for large-scale pumped-storage power station projects, the construction control network has high accuracy requirements. The selection of benchmark types should abandon the idea of simply obeying economy and pursuing simplicity. It is advisable to build observation piers with forced centering devices instead of ordinary benchmarks, otherwise it will directly affect the observation accuracy and work efficiency.
[0147] Second, in the application and selection of surveying and mapping instruments, the performance of the instruments and equipment must be stable and strong, and the accuracy indicators must be high. This is beneficial for field surveying on the one hand, and also provides hardware guarantee for accurate, uniform and reliable results on the other.
[0148] Third, the operators' excellent professional qualities and rich practical experience are basic conditions for obtaining qualified observation data and cannot be ignored.
[0149] The three-dimensional construction control networks of the above four projects were re-measured in the middle of the construction. Except for a few points that were damaged due to construction, they were all stable and reliable, and the range of change of their coordinate components was within the sensitivity range. Through the experimental research on the key technologies of the three-dimensional network, the methods of handling and solving related problems have achieved the expected purpose and achieved good application results. Relying on the research results and engineering experience of the key technologies of the three-dimensional network, in view of the special terrain conditions of the Fengning Pumped Storage Power Station, it is planned to use this technology to try to solve the difficulties and problems in the overall construction control of the power station project, and seek new application breakthroughs of this technology in network construction.
[0150] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be considered as the scope of protection of the present invention.
Claims
1. A three-dimensional control network adjustment method, characterized in that: The following steps are involved: Step S1, establishing a three-dimensional independent coordinate system: The center point is selected as the starting point of the control network in the control network, and the positive direction of the Z axis is the direction opposite to the vertical line direction of the starting point of the control network, which is the elevation direction; a certain normal high water level surface is selected as the projection surface of the plane coordinates, so as to establish a three-dimensional independent coordinate system; Step S2, determining five observation quantities, namely: horizontal angle observation quantity, slant distance observation quantity, zenith distance observation quantity, level height difference observation quantity and GNSS baseline observation quantity; For the control point i in the control network, relative to the sighting point k, one or more of the following five observation values are measured: horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik ; Step S3, perform a correction process on each observation value to obtain the correction number of each observation value: The horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik Perform the correction and get the horizontal angle correction value V Uik , slope distance correction number V Sik , Zenith distance correction V Bik , Level height difference correction factor V hik and GNSS baseline correction V Gik ; Step S4, determine the approximate value of each observation: Determine each horizontal angle observation value U separately ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The approximate value of the horizontal angle is obtained Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation Step S5, determine the adjustment value of each observation: Use formula (1) to formula (5) to determine each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The adjustment value of the horizontal angle is obtained Slope adjustment Zenith anomaly difference Level height difference adjustment value and GNSS baseline adjustment values Where: θ i is the azimuth of control point i, and are the approximate horizontal angles Slope range approximation Approximate zenith distance Approximate value of level height difference and GNSS baseline approximation number of corrections; Step S6, establish the error equation: Using formula (6) to formula (10), we can establish each horizontal angle observation value U ik , Slope range observation value S ik , zenith distance observation value B ik , level height difference observation value H ik and GNSS baseline observation value G ik The error equation is: Where: ρ is the atmospheric density; and are the deviations of the control point i and the aiming point k in the x-direction, y-direction and z-direction respectively; is the approximate value of horizontal distance; l Uik is the horizontal angle constant; l Sik is the slope distance constant term; l bik is the zenith distance constant term; l hik is the constant term of level height difference; l gik is the GNSS baseline constant term; x i ,y i and z i , is the three-dimensional coordinate of the control point i to be solved; x k ,y k and z k , is the three-dimensional coordinate of the aiming point k, which is a known value; Step S7, multiple control points are arranged in the control network, and the number of control points whose three-dimensional coordinates are to be solved is m1, so the number of unknown parameters to be solved is m=3m1; Count the number of observations of each type for all control points. Suppose a total of n1 horizontal angle observations U are obtained. ik , n2 slant range observations S ik , n3 zenith distance observations B ik , n4 level height difference observation values H ik , and n5 GNSS baseline observations G ik ; Therefore, the total number of observations is n = n1 + n2 + ... + n5; The matrix representation of the error equation group established in step S6 is obtained as follows: Where: V is a correction matrix with n rows and 1 column; A is a coefficient matrix with n rows and m columns; E is a three-dimensional coordinate matrix (x, y, z) with m rows and 1 column; L is a constant term matrix with n rows and 1 column; Step S8, solving the three-dimensional coordinates of each control point: Step S8.1, the observation weights of the horizontal angle observation, the slant distance observation, the zenith distance observation, the level height difference observation and the GNSS baseline observation are respectively expressed as: P1, P2, P3, P4, P5; the initial values of the observation weights P1, P2, P3, P4, P5 are given; Step S8.2, according to the matrix representation of the error equation group shown in formula (11), obtain the normal equation coefficient matrix N shown in formula (12) and the normal equation constant matrix W shown in formula (13): The equation coefficient matrix N is: Wherein, the superscript T represents the transformation rank of the matrix; A1, A2, A3, A4, A5 represent the coefficient matrix components of the coefficient matrix A regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively; N1, N2, N3, N4, N5 represent the coefficient matrix of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively; The equation constant matrix W is: Where: l1, l2, l3, l4, l5 represent the constant term components of the constant term matrix L regarding the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively; W1, W2, W3, W4, and W5 represent the constant arrays of horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation, and GNSS baseline observation, respectively; Step S8.3, the solution vector of the unknown parameters E = (X, Y, Z) is: Where: Q XX is the cofactor matrix; Thus, the three-dimensional coordinates of each control point are obtained; Step S8.4, using formula (14), obtain the unit weight variance of the adjustment value Among them, P = [P1, P2, P3, P4, P5] is the observation weight matrix; V = [V Uik ,V Sik ,V Bik ,V hik ,V Gik ], is the observation correction matrix; r is the total number of redundant observations; Step S8.5, using formula (15), calculate the posterior variance of each observation j Among them, j = 1, 2, 3, 4, 5, representing the horizontal angle observation, slant distance observation, zenith distance observation, level height difference observation and GNSS baseline observation respectively; Where: P j represents the observation weight of observation j; V j represents the correction number of observation j; R j is the number of redundant observations of observation j; Step S8.6, determine the posterior variance of each observation j Is the deviation from the prior variance of each observation j within the set range? If so, end the process and output the three-dimensional coordinates of each control point solved by step S8.3; otherwise, adjust the observation weights P1, P2, P3, P4, P5 of each observation and return to step S8.2 to loop.
2. A three-dimensional control network adjustment method according to claim 1, characterized in that: In step S3, the horizontal angle correction value V Uik 0; GNSS baseline correction number V Gik is 0; For the slope distance observation value S ik , respectively make corrections for ruler length, meteorological correction, and channel curvature, and then return to the mark center; For the zenith distance observation value B ik , respectively, the zenith distance observation value B ik Correction normalized to the center of the mark and spherical aberration correction; For the level height difference observation value H ik , if H ik If the distance is greater than 500m, corrections are made for the length of the rod, non-parallelism of the level surface and gravity anomaly.
3. A three-dimensional control network adjustment method according to claim 1, characterized in that: In step S3 and step S4, the following method is used to obtain the approximate value of each observation value: The control network points for which approximate values need to be obtained are jointly measured with the plane coordinates and elevations of the known points. Each known point is used as a joint measurement position. Therefore, there are no less than three joint measurement positions, and they are evenly distributed in the entire survey area. Then a similarity transformation is performed to make the sum of the squares of the residuals of the coordinate components of each point extremely small, so as to obtain various approximate values of the control network points.
Citation Information
Patent Citations
Rigorous three-dimensional adjustment method for edge-angle intersection net of free observation station
CN106980598A
Method and device for processing observation data of control points in three-dimensional control network
CN110793508A
Long tunnel multi-observation-station combined measurement combination resolving method and device and storage medium
CN114877860A
Multi-reference-station-network-based GNSS baseline joint calculation method and computer readable medium
CN116882134A
