Intelligent leveling and hoisting multi-modal control method and system for turbulent flow adaptive steel casing

Abstract

The application discloses a kind of turbulent flow adaptive steel casing intelligent leveling hoisting multi-modal control method and system, the method includes: in steel casing hoisting operation area, multiple-source sensor is laid out, and the original signal of each sensor is collected in real time, the observation of each sensor is obtained by normalizing the original signal, according to the preset multi-modal perception fusion function, the observation of each sensor is fused into fusion perception field;The corresponding frequency domain field is obtained by Fourier transform to fusion perception field, and the probability of frequency domain field frequency energy in each spectral energy box is calculated to calculate local spectral entropy, according to frequency domain field and local spectral entropy, turbulent flow adaptive frequency spectrum is calculated;The current attitude error vector of steel casing is mapped to by mapping operator through turbulent flow adaptive frequency spectrum, and according to target leveling attitude vector, tracking weight matrix and energy consumption balance penalty term, control vector is solved, and leveling instruction of each lifting point is generated by anti-normalization to control vector.

Description

Technical Field

[0001] This invention belongs to the field of steel casing hoisting technology, and more specifically, relates to a multimodal control method and system for intelligent leveling and hoisting of turbulent adaptive steel casing. Background Technology

[0002] In existing large steel casing hoisting technologies, leveling is typically achieved through manual operation or semi-automatic control. The main technical solutions include the following aspects: First, the perception of the steel casing's attitude and environmental disturbances during hoisting relies heavily on single sensors, such as single-point inclinometers or lifting point load sensors, lacking multi-source information fusion capabilities. Second, existing control strategies mostly employ fixed proportional, PID, or linear feedback regulation, failing to dynamically adapt to changes in turbulence and wind load in the field environment. This leads to the steel casing easily experiencing horizontal shifts, tilting, or even oscillations under turbulent or sudden wind load changes, increasing hoisting safety risks. Third, existing methods have limited ability to suppress high-frequency disturbances. For short-term high-frequency vibrations caused by turbulence, the controller cannot perform real-time filtering, resulting in attitude adjustment lag, overshoot, or oscillation, further reducing leveling accuracy.

[0003] Therefore, there is an urgent need for a technical solution that can enable steel casings to maintain high precision and stable posture, significantly improve hoisting safety and operational efficiency, while taking into account energy consumption optimization and structural balance, and achieve intelligent, closed-loop, and robust hoisting leveling control. Summary of the Invention

[0004] To address the above technical problems, this invention proposes a multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing, comprising:

[0005] Multi-source sensors are deployed in the steel casing hoisting operation area, and the raw signals of each sensor are collected in real time. The raw signals are normalized to obtain the observations of each sensor. According to the preset multimodal sensing fusion function, the observations of each sensor are fused into a fused sensing field.

[0006] The fused sensing field is subjected to Fourier transform to obtain the corresponding frequency domain field, and the local spectral entropy is calculated based on the probability of the frequency domain energy in each spectral energy box according to the frequency domain energy of the frequency domain field. The turbulence adaptive spectrum is calculated based on the frequency domain field and the local spectral entropy.

[0007] The turbulence adaptive spectrum is mapped to the current attitude error vector of the steel casing by the mapping operator. The control vector is solved based on the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0008] Furthermore, the multimodal sensing fusion function includes:

[0009] ,

[0010] in, For time Time position The fusion perception field at the location, For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0011] Furthermore, calculating the local spectral entropy includes:

[0012] ,

[0013] in, For time Time position frequency domain wavenumber Local spectral entropy at that location The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First Normalized probability of each spectral energy box To avoid tiny positive numbers in log(0).

[0014] Furthermore, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0015] ,

[0016] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, As an adaptive inhibitory factor, For time Time position frequency domain wavenumber The frequency domain field at that location.

[0017] Furthermore, solving for the control vector includes:

[0018] ,

[0019] in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

[0020] Furthermore, acquiring time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

[0021] This invention also proposes a multi-modal control system for intelligent leveling and hoisting of turbulent adaptive steel casing, comprising:

[0022] The fusion module is used to deploy multi-source sensors in the steel casing hoisting operation area, collect the raw signals of each sensor in real time, normalize the raw signals to obtain the observations of each sensor, and fuse the observations of each sensor into a fused sensing field according to the preset multimodal sensing fusion function.

[0023] The calculation module is used to perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy.

[0024] The leveling module is used to map the turbulent adaptive spectrum into the current attitude error vector of the steel casing through the mapping operator, and solve the control vector based on the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0025] Furthermore, the multimodal sensing fusion function includes:

[0026] ,

[0027] in, For time Time position The fusion perception field at the location, For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0028] Furthermore, calculating the local spectral entropy includes:

[0029] ,

[0030] in, For time Time position frequency domain wavenumber Local spectral entropy at that location The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First Normalized probability of each spectral energy box To avoid tiny positive numbers in log(0).

[0031] Furthermore, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0032] ,

[0033] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, As an adaptive inhibitory factor, For time Time position frequency domain wavenumber The frequency domain field at that location.

[0034] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0035] This invention achieves real-time perception and dynamic compensation of environmental turbulence during the hoisting of steel casings through multimodal scene perception, turbulence adaptive frequency domain filtering, and multimodal adaptive leveling control allocation. This enables the steel casings to maintain high precision level and stable posture, significantly improving hoisting safety and operational efficiency. At the same time, it takes into account energy consumption optimization and structural balance, realizing intelligent, closed-loop, and robust hoisting leveling control. Attached Figure Description

[0036] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention;

[0037] Figure 2 This is a system structure diagram of Embodiment 2 of the present invention. Detailed Implementation

[0038] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0039] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0040] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0041] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0042] The display screen is used to show the user interface of each application.

[0043] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0044] Example 1

[0045] like Figure 1As shown in the figure, this embodiment proposes a multimodal control method for intelligent leveling and hoisting of turbulent adaptive steel casing, including:

[0046] Step 101: Deploy multi-source sensors in the steel casing hoisting operation area and collect the raw signals of each sensor in real time. Normalize the raw signals to obtain the observations of each sensor. According to the preset multimodal sensing fusion function, fuse the observations of each sensor into a fused sensing field.

[0047] Preferably, the multi-source sensors can be: lidar: measuring the three-dimensional displacement information of the top of the casing; accelerometer: measuring the acceleration of the casing center to identify dynamic disturbances; wind speed sensor: collecting local wind speed and wind direction information, etc.

[0048] Specifically, multimodal sensing fusion functions include:

[0049] ,

[0050] in, For time Time position (Both time and location need to be normalized) the fused perception field at the location. For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0051] Preferred, Specifically:

[0052] ,

[0053] in, To prevent dividing by zero and positive numbers.

[0054] Step 102: Perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy.

[0055] Specifically, calculating the local spectral entropy includes:

[0056] ,

[0057] in, For time Time position frequency domain wavenumber The local spectral entropy at a given point represents the wavenumber in the frequency domain. The degree of disorder in the energy distribution of the spectrum (the higher the degree, the stronger the degree of turbulence chaos). The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First The normalized probability of each spectral energy bin (can be obtained from the short-time spectral energy allocation within a local window). To avoid tiny positive numbers in log(0).

[0058] Specifically, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0059] ,

[0060] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, Adaptive inhibitory factor ( (greater than or equal to 0 and less than 1) For time Time position frequency domain wavenumber The frequency domain field at that location.

[0061] Step 103: The turbulence adaptive spectrum is mapped to the current attitude error vector of the steel casing by the mapping operator. The control vector is solved according to the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0062] Specifically, solving for the control vector includes:

[0063] ,

[0064] in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

[0065] Specifically, acquisition time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

[0066] Preferably, the total risk indicators for each lifting point include:

[0067] ,

[0068] in, For time Time Risk of deviation at each lifting point For time Time The actual displacement of each lifting point For the first The reference lifting point displacement of each lifting point For the first The maximum allowable displacement of each lifting point.

[0069] ,

[0070] in, For time Time Structural constraint risks at each suspension point For time Time The actual tension at each suspension point For safety tension, For time Time The maximum permissible tension at each suspension point.

[0071] ,

[0072] in, For time Time The overall risk index of each lifting point As the weight of deviation risk, The weight of structural constraint risk.

[0073] Calculate the tracking weight for each lifting point:

[0074] ,

[0075] in, For time Time The tracking weight of each suspension point To adjust the factors, control the steepness of the curve, and regulate the sensitivity to changes in weights, This is the risk threshold.

[0076] ,

[0077] in, To generate diagonal matrix operators, This refers to the number of lifting points.

[0078] Example 2

[0079] like Figure 2 As shown, this embodiment proposes a multi-modal control system for intelligent leveling and hoisting of turbulent adaptive steel casing, including:

[0080] The fusion module is used to deploy multi-source sensors in the steel casing hoisting operation area, collect the raw signals of each sensor in real time, normalize the raw signals to obtain the observations of each sensor, and fuse the observations of each sensor into a fused sensing field according to the preset multimodal sensing fusion function.

[0081] Preferably, the multi-source sensors can be: lidar: measuring the three-dimensional displacement information of the top of the casing; accelerometer: measuring the acceleration of the casing center to identify dynamic disturbances; wind speed sensor: collecting local wind speed and wind direction information, etc.

[0082] Specifically, multimodal sensing fusion functions include:

[0083] ,

[0084] in, For time Time position (Both time and location need to be normalized) the fused perception field at the location. For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0085] Preferred, Specifically:

[0086] ,

[0087] in, To prevent dividing by zero and positive numbers.

[0088] The calculation module is used to perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy.

[0089] Specifically, calculating the local spectral entropy includes:

[0090] ,

[0091] in, For time Time position frequency domain wavenumber The local spectral entropy at a given point represents the wavenumber in the frequency domain. The degree of disorder in the energy distribution of the spectrum (the higher the degree, the stronger the degree of turbulence chaos). The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First The normalized probability of each spectral energy bin (can be obtained from the short-time spectral energy allocation within a local window). To avoid tiny positive numbers in log(0).

[0092] Specifically, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0093] ,

[0094] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, Adaptive inhibitory factor ( (greater than or equal to 0 and less than 1) For time Time position frequency domain wavenumber The frequency domain field at that location.

[0095] The leveling module is used to map the turbulent adaptive spectrum into the current attitude error vector of the steel casing through the mapping operator, and solve the control vector based on the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0096] Specifically, solving for the control vector includes:

[0097] ,

[0098] in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

[0099] Specifically, acquisition time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

[0100] Preferably, the total risk indicators for each lifting point include:

[0101] ,

[0102] in, For time Time Risk of deviation at each lifting point For time Time The actual displacement of each lifting point For the first The reference lifting point displacement of each lifting point For the first The maximum allowable displacement of each lifting point.

[0103] ,

[0104] in, For time Time Structural constraint risks at each suspension point For time Time The actual tension at each suspension point For safety tension, For time Time The maximum permissible tension at each suspension point.

[0105] ,

[0106] in, For time Time The overall risk index of each lifting point As the weight of deviation risk, The weight of structural constraint risk.

[0107] Calculate the tracking weight for each lifting point:

[0108] ,

[0109] in, For time Time The tracking weight of each suspension point To adjust the factors, control the steepness of the curve, and regulate the sensitivity to changes in weights, This is the risk threshold.

[0110] ,

[0111] in, To generate diagonal matrix operators, This refers to the number of lifting points.

[0112] Example 3

[0113] This invention also proposes a storage medium storing multiple instructions, which are used to implement the aforementioned turbulence adaptive steel casing intelligent leveling and hoisting multimodal control method.

[0114] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0115] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following method steps: Step 101, deploy multi-source sensors in the steel casing hoisting operation area and collect the original signals of each sensor in real time, normalize the original signals to obtain the observations of each sensor, and fuse the observations of each sensor into a fused sensing field according to a preset multimodal sensing fusion function.

[0116] Preferably, the multi-source sensors can be: lidar: measuring the three-dimensional displacement information of the top of the casing; accelerometer: measuring the acceleration of the casing center to identify dynamic disturbances; wind speed sensor: collecting local wind speed and wind direction information, etc.

[0117] Specifically, multimodal sensing fusion functions include:

[0118] ,

[0119] in, For time Time position (Both time and location need to be normalized) the fused perception field at the location. For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0120] Preferred, Specifically:

[0121] ,

[0122] in, To prevent dividing by zero and positive numbers.

[0123] Step 102: Perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy.

[0124] Specifically, calculating the local spectral entropy includes:

[0125] ,

[0126] in, For time Time position frequency domain wavenumber The local spectral entropy at a given point represents the wavenumber in the frequency domain. The degree of disorder in the energy distribution of the spectrum (the higher the degree, the stronger the degree of turbulence chaos). The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First The normalized probability of each spectral energy bin (can be obtained from the short-time spectral energy allocation within a local window). To avoid tiny positive numbers in log(0).

[0127] Specifically, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0128] ,

[0129] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, Adaptive inhibitory factor ( (greater than or equal to 0 and less than 1) For time Time position frequency domain wavenumber The frequency domain field at that location.

[0130] Step 103: The turbulence adaptive spectrum is mapped to the current attitude error vector of the steel casing by the mapping operator. The control vector is solved according to the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0131] Specifically, solving for the control vector includes:

[0132] ,

[0133] in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

[0134] Specifically, acquisition time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

[0135] Preferably, the total risk indicators for each lifting point include:

[0136] ,

[0137] in, For time Time Risk of deviation at each lifting point For time Time The actual displacement of each lifting point For the first The reference lifting point displacement of each lifting point For the first The maximum allowable displacement of each lifting point.

[0138] ,

[0139] in, For time Time Structural constraint risks at each suspension point For time Time The actual tension at each suspension point For safety tension, For time Time The maximum permissible tension at each suspension point.

[0140] ,

[0141] in, For time Time The overall risk index of each lifting point As the weight of deviation risk, The weight of structural constraint risk.

[0142] Calculate the tracking weight for each lifting point:

[0143] ,

[0144] in, For time Time The tracking weight of each suspension point To adjust the factors, control the steepness of the curve, and regulate the sensitivity to changes in weights, This is the risk threshold.

[0145] ,

[0146] in, To generate diagonal matrix operators, This refers to the number of lifting points.

[0147] Example 4

[0148] This invention also proposes an electronic device, including a processor and a storage medium connected to the processor. The storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to execute the aforementioned turbulence adaptive steel casing intelligent leveling and hoisting multimodal control method.

[0149] Specifically, the electronic device in this embodiment can be a computer terminal, which may include one or more processors and a storage medium.

[0150] The storage medium can be used to store software programs and modules, such as the multimodal control method for intelligent leveling and hoisting of turbulent adaptive steel casing in this embodiment of the invention. The processor executes the software programs and modules stored in the storage medium to perform various functional applications and data processing, thus realizing the aforementioned multimodal control method for intelligent leveling and hoisting of turbulent adaptive steel casing. The storage medium may include high-speed random access storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely configured relative to the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0151] The processor can call the information and application stored in the storage medium through the transmission system to execute the following steps: Step 101, deploy multi-source sensors in the steel casing hoisting operation area and collect the original signals of each sensor in real time, normalize the original signals to obtain the observations of each sensor, and fuse the observations of each sensor into a fused sensing field according to the preset multimodal sensing fusion function.

[0152] Preferably, the multi-source sensors can be: lidar: measuring the three-dimensional displacement information of the top of the casing; accelerometer: measuring the acceleration of the casing center to identify dynamic disturbances; wind speed sensor: collecting local wind speed and wind direction information, etc.

[0153] Specifically, multimodal sensing fusion functions include:

[0154] ,

[0155] in, For time Time position (Both time and location need to be normalized) the fused perception field at the location. For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

[0156] Preferred, Specifically:

[0157] ,

[0158] in, To prevent dividing by zero and positive numbers.

[0159] Step 102: Perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy.

[0160] Specifically, calculating the local spectral entropy includes:

[0161] ,

[0162] in, For time Time position frequency domain wavenumber The local spectral entropy at a given point represents the wavenumber in the frequency domain. The degree of disorder in the energy distribution of the spectrum (the higher the degree, the stronger the degree of turbulence chaos). The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First The normalized probability of each spectral energy bin (can be obtained from the short-time spectral energy allocation within a local window). To avoid tiny positive numbers in log(0).

[0163] Specifically, the calculation of the adaptive turbulence spectrum based on the frequency domain field and local spectral entropy includes:

[0164] ,

[0165] in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, Adaptive inhibitory factor ( (greater than or equal to 0 and less than 1) For time Time position frequency domain wavenumber The frequency domain field at that location.

[0166] Step 103: The turbulence adaptive spectrum is mapped to the current attitude error vector of the steel casing by the mapping operator. The control vector is solved according to the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

[0167] Specifically, solving for the control vector includes:

[0168] ,

[0169] in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

[0170] Specifically, acquisition time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

[0171] Preferably, the total risk indicators for each lifting point include:

[0172] ,

[0173] in, For time Time Risk of deviation at each lifting point For time Time The actual displacement of each lifting point For the first The reference lifting point displacement of each lifting point For the first The maximum allowable displacement of each lifting point.

[0174] ,

[0175] in, For time Time Structural constraint risks at each suspension point For time Time The actual tension at each suspension point For safety tension, For time Time The maximum permissible tension at each suspension point.

[0176] ,

[0177] in, For time Time The overall risk index of each lifting point As the weight of deviation risk, The weight of structural constraint risk.

[0178] Calculate the tracking weight for each lifting point:

[0179] ,

[0180] in, For time Time The tracking weight of each suspension point To adjust the factors, control the steepness of the curve, and regulate the sensitivity to changes in weights, This is the risk threshold.

[0181] ,

[0182] in, To generate diagonal matrix operators, This refers to the number of lifting points.

[0183] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0184] In the several embodiments provided by this invention, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0185] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0186] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0187] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, optical disks, and other media capable of storing program code.

[0188] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A multimodal control method for intelligent leveling and hoisting of turbulent adaptive steel casing, characterized in that, include: Multi-source sensors are deployed in the steel casing hoisting operation area, and the raw signals of each sensor are collected in real time. The raw signals are normalized to obtain the observations of each sensor. According to the preset multimodal sensing fusion function, the observations of each sensor are fused into a fused sensing field. The fused sensing field is subjected to Fourier transform to obtain the corresponding frequency domain field, and the local spectral entropy is calculated based on the probability of the frequency domain energy in each spectral energy box according to the frequency domain energy of the frequency domain field. The turbulence adaptive spectrum is calculated based on the frequency domain field and the local spectral entropy. The turbulence adaptive spectrum is mapped to the current attitude error vector of the steel casing by the mapping operator. The control vector is solved based on the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

2. The multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing as described in claim 1, characterized in that, Multimodal sensing fusion functions include: ， in, For time Time position The fusion perception field at the location, For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

3. The multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing as described in claim 2, characterized in that, Calculating the local spectral entropy includes: ， in, For time Time position frequency domain wavenumber Local spectral entropy at that location The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First Normalized probability of each spectral energy box To avoid tiny positive numbers in log(0).

4. The multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing as described in claim 3, characterized in that, The calculation of the adaptive spectrum of turbulence based on the frequency domain field and local spectral entropy includes: ， in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, As an adaptive inhibitory factor, For time Time position frequency domain wavenumber The frequency domain field at that location.

5. The multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing as described in claim 4, characterized in that, Solving for the control vector includes: ， in, For time Time position The control vector at that location, Candidate control vectors, For the candidate control set, For time Time tracking weight matrix, It is a mapping operator, and it is the identity matrix. For frequency domain inverse transform, For time Time position The target leveling attitude vector at the location.

6. The multi-modal control method for intelligent leveling and hoisting of turbulent adaptive steel casing as described in claim 5, characterized in that, Acquisition Time Time tracking weight matrix Includes: computation time For each lifting point, calculate the risk index, then sum the risk indices of each lifting point using weighted averages to obtain the total risk index for each lifting point. Finally, construct a diagonal matrix from the total risk indices of all lifting points to represent the time period. Time tracking weight matrix .

7. A multi-modal control system for intelligent leveling and hoisting of turbulent adaptive steel casing, characterized in that, include: The fusion module is used to deploy multi-source sensors in the steel casing hoisting operation area, collect the raw signals of each sensor in real time, normalize the raw signals to obtain the observations of each sensor, and fuse the observations of each sensor into a fused sensing field according to the preset multimodal sensing fusion function. The calculation module is used to perform Fourier transform on the fused sensing field to obtain the corresponding frequency domain field, calculate the local spectral entropy based on the probability of the frequency domain energy in each spectral energy box, and calculate the turbulence adaptive spectrum based on the frequency domain field and the local spectral entropy. The leveling module is used to map the turbulent adaptive spectrum into the current attitude error vector of the steel casing through the mapping operator, and solve the control vector based on the target leveling attitude vector, the tracking weight matrix and the energy consumption balance penalty term. The control vector is then denormalized to generate leveling commands for each lifting point.

8. The turbulence adaptive steel casing intelligent leveling and hoisting multimodal control system as described in claim 7, characterized in that, Multimodal sensing fusion functions include: ， in, For time Time position The fusion perception field at the location, For the number of sensors, For the first The weight of each sensor, For time Time position First The observations of each sensor, For time Time The average of the observations from each sensor, This is the regularization kernel function. For time Time The standard deviation of observations of each sensor.

9. The turbulence adaptive steel casing intelligent leveling and hoisting multimodal control system as described in claim 8, characterized in that, Calculating the local spectral entropy includes: ， in, For time Time position frequency domain wavenumber Local spectral entropy at that location The number of spectral energy boxes, For time Wavenumber in the frequency domain Location First Normalized probability of each spectral energy box To avoid tiny positive numbers in log(0).

10. The turbulence adaptive steel casing intelligent leveling and hoisting multimodal control system as described in claim 9, characterized in that, The calculation of the adaptive spectrum of turbulence based on the frequency domain field and local spectral entropy includes: ， in, For time Time position frequency domain wavenumber Turbulence adaptive spectrum at the location, As an adaptive inhibitory factor, For time Time position frequency domain wavenumber The frequency domain field at that location.

Images