Electric control method and system for double-tire beam carrying machine based on intelligent redundancy switching

Abstract

The invention discloses a double-tire beam moving machine electrical control method based on intelligent redundancy switching, and the method comprises the steps: obtaining position data through a GNSS system and an RTK system at the same time, and obtaining environment data through an environment sensor unit; inputting each data input data into an LSTM-VAE model, and selecting optimal measurement source data; calculating positioning information; and synchronous displacement control of straight movement and / or transverse movement of the double tires is carried out. The invention also discloses a system based on the method. The system comprises a positioning data acquisition unit, a walking synchronous control unit, a supporting leg inclination and stress monitoring unit, a tire pressure detection unit, a steel wire rope detection unit, a main beam deflection monitoring unit and a remote monitoring unit. According to the electrical control method and system for the double-tire type beam carrying machine based on intelligent redundancy switching, the multiple control units are arranged and cooperate with one another, and therefore the adaptability and efficiency of the beam carrying machine in high-speed railway construction are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of box girder erection beam moving machines, and in particular to an electrical control method and system for a dual-tire beam moving machine based on intelligent redundant switching. Background Art

[0002] In the field of high-speed railway and bridge engineering construction, traditional dual-tire beam movers have a series of challenges in operating performance, which limits their effective use in high-speed railway construction. Traditional beam movers are usually limited by limited automation levels and lack of monitoring of key operating parameters, resulting in low operating efficiency and potential safety hazards during construction. Traditional beam movers have limited walking synchronization control, lack of high-precision positioning and intelligent navigation functions, and are almost completely dependent on human experience control, which may lead to inaccurate position control and operational difficulties in high-speed railway and bridge construction. If automatic positioning measurement technology can be introduced into the electrical control of dual-tire beam movers, the degree of automation control can be improved.

[0003] Existing positioning methods include GNSS (Global Navigation Satellite System) and RTK (Real Time Kinematic). GNSS technology performs well in scenarios with relatively low positioning accuracy requirements such as outdoor navigation and location services, especially when satellites can be clearly seen in the sky over open areas. However, when atmospheric conditions change significantly, such as fluctuations in atmospheric humidity and temperature, signal refraction and delay may occur, affecting positioning accuracy. For tasks that require high-precision measurements, RTK technology can provide more accurate and real-time positioning information. However, in RTK, when the baseline length (the distance between the receiver and the reference station) is long, larger measurement errors may result.

[0004] How to flexibly switch between GNSS and RTK for different scenario tasks, provide accurate position information basis for the electrical control of the beam moving machine, and make the beam moving machine more intelligent and automated, is a technical problem that needs to be solved urgently.

[0005] In addition, the lack of comprehensive autonomous driving and anti-collision mechanisms makes the beam mover lack safety and autonomy, which may lead to collision risks and unstable operation. The problem of monitoring the inclination and stress of the outriggers is also one of the challenges of traditional beam movers. Insufficient accurate monitoring of the inclination and stress of the outriggers may lead to an unsafe working environment and uneven structural stress. In addition, the lack of a remote monitoring system makes it difficult to remotely monitor and manage the real-time operating status of the beam mover, limiting the operator's comprehensive understanding of the equipment. Monitoring of key parameters such as tire pressure, wire rope, and main beam deflection is also a problem with traditional beam movers, which also limits the comprehensive evaluation of the structure and performance of the beam mover. Summary of the invention

[0006] In order to address the deficiencies in the prior art, the present invention provides an electrical control method and system for a dual-tyre beam moving machine based on intelligent redundant switching, wherein a plurality of control units are established and the units cooperate with each other, thereby significantly improving the adaptability and efficiency of the beam moving machine in high-speed railway construction.

[0007] In a first aspect, an embodiment of the present invention provides an electrical control method for a dual-tyre beam moving machine based on intelligent redundant switching, comprising the following process:

[0008] Step 1: Use the GNSS system and the RTK system to simultaneously acquire location data, and use the environmental sensor unit to acquire environmental data;

[0009] Step 2: Input the location data and environmental data of the GNSS system and / or RTK system as input data into the LSTM-VAE model for potential data decision processing, and select the optimal measurement source data;

[0010] Step 3: Calculate positioning information using optimal measurement source data;

[0011] Step 4: The running synchronization control unit performs synchronous control of the straight and / or lateral displacement of the dual tires based on the positioning information.

[0012] In an optional embodiment, the environmental data described in step 2 includes weather conditions, terrain information, and surrounding building information.

[0013] In an optional embodiment, the sensor information data described in step 2 includes the signal-to-noise ratio, positioning error and signal quality of the sensor.

[0014] In an optional embodiment, step 2 specifically includes the following process:

[0015] S2.1. Convert the location data and environmental data of the GNSS system and / or RTK system into an input sequence and input it into the LSTM network to obtain the LSTM hidden state:

[0016] h t =LSTM(x t ,h t-1 )

[0017] Among them, x t is the position data and environmental data of the GNSS system and / or RTK system at time t, h t is the hidden state of LSTM at time t;

[0018] S2.2. The encoder is used to map the LSTM hidden state to the mean and variance parameters of the latent space, expressed as:

[0019] μ,σ 2 =Encoder(h t )

[0020] Among them, μ and σ 2 denote the mean and variance of the latent space respectively;

[0021] A latent variable z is sampled from the mean and variance of the encoder output, expressed as:

[0022] z~N(μ,σ 2 )

[0023] Where N represents the number of samples;

[0024] S2.3. Use the decoder to decode the latent variable z into a reconstruction of the input sequence, expressed as:

[0025] x t '=Decoder(z)

[0026] Among them, x t ' is to reconstruct sequence data;

[0027] S2.4. Calculate the reconstruction loss and KL divergence loss. The reconstruction loss is the difference between the reconstructed sequence data and the input sequence, and is calculated using the mean square error function:

[0028]

[0029] Among them, Reconstruction Loss represents the reconstruction loss;

[0030] The KL divergence loss is the difference between the distribution of the latent variable and the standard normal distribution and is calculated using the following formula:

[0031]

[0032] Among them, KL DivergenceLoss represents the KL divergence loss, and J is the dimension of the latent space;

[0033] S2.5. Train through back propagation and gradient descent algorithms to minimize reconstruction loss and KL divergence loss, obtain the optimal solution, output the reconstructed sequence data, and form the best measurement source data.

[0034] In an optional embodiment, the input sequence described in step S2.1 is obtained through dynamic selection processing of the measurement source, including the following process:

[0035] a1. According to the current environmental factors and sensor information data, the location data and environmental data of the GNSS system and RTK system are scored respectively;

[0036] a2. Take the location data and environmental data corresponding to the source with the high score as the input sequence.

[0037] In an optional embodiment, the displacement synchronization control described in step 4 includes:

[0038] The Hall speed sensor installed on each electric motor is used to collect the walking speed data, and the PID control is used to perform closed-loop speed regulation;

[0039] The laser ranging sensor is used to monitor the displacement data of the two tires at the same time, and the displacement data of the two tires are synchronized in combination with the positioning information to perform displacement control.

[0040] In the second aspect, an embodiment of the present invention provides an electrical control system for a dual-tire beam moving machine based on the method, comprising a positioning data acquisition unit, a running synchronization control unit, a leg tilt and stress monitoring unit, a tire pressure detection unit, a wire rope detection unit, a main beam deflection monitoring unit and a remote monitoring unit connected in a CAN bus networking manner, wherein the positioning data acquisition unit comprises a GNSS system and an RTK system.

[0041] In an optional embodiment, the running synchronization control unit includes a Hall speed sensor and a laser ranging sensor installed at each motor, and a Hall speed sensor and a laser ranging sensor for collecting data.

[0042] In an optional embodiment, the leg tilt and stress monitoring unit includes an angle tilter installed on the leg and a vibrating wire strain gauge installed inside the mold frame.

[0043] In an optional embodiment, the autonomous driving intelligent collision avoidance unit includes an inertial navigation system, a camera, and a radar sensor.

[0044] The beneficial effects of the present invention based on its technical solution are:

[0045] (1) The electrical control method and system of a dual-tire beam moving machine based on intelligent redundant switching provided by the present invention utilizes the LSTM-VAE model to process potential data decisions. The long short-term memory network (LSTM) and the variational autoencoder (VAE) are two different types of neural network models, which are used to process sequence data and generate representations of latent space, respectively. Combining them can create a powerful model that gives full play to their respective advantages and dynamically selects the optimal measurement source according to the real-time environment and sensor performance. For example, when the GNSS signal is interfered with, the system can automatically switch to the RTK measurement technology to maintain high-precision measurement. When the RTK signal is interfered with, the system can automatically switch to the GNSS measurement technology to maintain high-precision measurement, so as to fully utilize their respective advantages to meet the different requirements of specific scenarios for measurement accuracy, real-time performance, and robustness.

[0046] (2) The electrical control method and system of a dual-tire beam mover based on intelligent redundant switching provided by the present invention significantly improves the operating performance and automation level of the dual-tire beam mover by introducing key control modules such as a walking synchronization control unit, an automatic driving and an anti-collision unit. The walking synchronization control unit ensures the stability and controllability of the beam mover during its walking process, while the automatic driving and anti-collision unit enable the beam mover to have the ability of autonomous navigation and collision avoidance, thereby improving the flexibility and efficiency of construction.

[0047] (3) The remote monitoring unit integrated in the electrical control method and system of the dual-tire beam moving machine based on intelligent redundant switching provided by the present invention enables the real-time working data of the beam moving machine to be presented on the cloud platform in the form of digital twins, so that the operator can remotely monitor the status of the equipment in real time. This not only improves the safety of high-speed railway projects, but also enables operators to respond to any potential problems in a timely manner and reduce the risk of accidents.

[0048] (3) The electrical control method and system of a dual-tire beam moving machine based on intelligent redundant switching provided by the present invention integrates multiple acquisition and control units, and comprehensively monitors the key parameters of the transporter through the comprehensive application of multiple monitoring units such as the outrigger tilt and stress monitoring unit, the tire pressure monitoring unit, the wire rope detection unit, and the main beam deflection monitoring unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1A structural block diagram of an electrical control system of a dual-tyre beam moving machine provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention belong to the scope of protection of the embodiments of the present invention.

[0052] The embodiment of the present invention provides an electrical control method for a dual-tire beam moving machine based on intelligent redundant switching, comprising the following process:

[0053] Step 1: Use the GNSS system and the RTK system to simultaneously acquire location data, and use the environmental sensor unit to acquire environmental data.

[0054] Step 2. LSTM (long short-term memory) and VAE (Variational Auto-Encoders) are two different types of neural network models, which are used to process sequence data and generate representations of latent space respectively. Combining them can create a powerful model that gives full play to their respective advantages. First, the LSTM network can be used to model the input sequence data. LSTM can effectively capture long-term dependencies in sequence data. In this algorithm, LSTM will be used to model sensor data in order to better understand the dynamic changes of the environment.

[0055] The position data and environmental data of the GNSS system and / or RTK system are input into the LSTM-VAE model as input data for potential data decision processing to select the optimal measurement source data.

[0056] Environmental information and sensor performance information can be added to the input of the model. Environmental information includes factors such as the current weather conditions, terrain information, and surrounding buildings, which may affect the performance of different measurement sources. Sensor performance information includes performance indicators such as the current sensor's signal-to-noise ratio (SNR), positioning error, and signal quality. This information can help the model evaluate the applicability of different measurement sources.

[0057] Step 2 specifically includes the following processes:

[0058] S2.1. Convert the location data and environmental data of the GNSS system and / or RTK system into an input sequence and input it into the LSTM network to obtain the LSTM hidden state:

[0059] h t=LSTM(x t ,h t-1 )

[0060] Among them, x t is the position data and environmental data of the GNSS system and / or RTK system at time t, h t is the hidden state of the LSTM at time t. The initial state at time t = 0 is usually called the initial hidden state. By default, the initial hidden state is initialized to an all-zero vector. The input sequence is obtained by dynamic selection of the measurement source, including the following process:

[0061] a1. Score the location data and environmental data of the GNSS system and RTK system respectively based on the current environmental factors (such as weather conditions, terrain information and surrounding building information) and sensor information data (such as signal-to-noise ratio, positioning error and signal quality);

[0062] a2. Take the location data and environmental data corresponding to the source with the high score as the input sequence.

[0063] Using the above evaluation model, a score is calculated for each measurement source (e.g., 0 to 10, where 10 represents the best performance and 0 represents the worst performance). A measurement source with a high score means that its data has high accuracy and reliability. Based on the calculated score, the measurement source with the highest score is selected as the optimal measurement source. This ensures that the system uses the most accurate and reliable data, which not only allows for a more systematic evaluation and selection of the optimal measurement source, but also ensures that the accuracy and stability of the entire system are further improved. The implementation of this step will make the system more intelligent and able to adapt to changing environments and different sensor performance, thereby improving overall operational efficiency and safety.

[0064] S2.2. The encoder is used to map the LSTM hidden state to the mean and variance parameters of the latent space, expressed as:

[0065] μ,σ 2 =Encoder(h t )

[0066] Among them, μ and σ 2 denote the mean and variance of the latent space respectively;

[0067] A latent variable z is sampled from the mean and variance of the encoder output, expressed as:

[0068] z~N(μ,σ 2 )

[0069] Where N represents the number of samples;

[0070] S2.3. Use the decoder to decode the latent variable z into a reconstruction of the input sequence, expressed as:

[0071] x t '=Decoder(z)

[0072] Among them, x t ' is to reconstruct sequence data;

[0073] S2.4. Calculate the reconstruction loss and KL divergence loss. The reconstruction loss is the difference between the reconstructed sequence data and the input sequence, and is calculated using the mean square error function:

[0074]

[0075] Among them, Reconstruction Loss represents the reconstruction loss;

[0076] The KL divergence loss is the difference between the distribution of the latent variable and the standard normal distribution and is calculated using the following formula:

[0077]

[0078] Among them, KLDivergenceLoss represents the KL divergence loss, and J is the dimension of the latent space;

[0079] S2.5. Train through back propagation and gradient descent algorithms to minimize reconstruction loss and KL divergence loss, obtain the optimal solution, output the reconstructed sequence data, and form the best measurement source data.

[0080] Step 3: Calculate positioning information using the optimal measurement source data.

[0081] Step 4: The running synchronization control unit performs the displacement synchronization control of the dual tires in the straight and / or transverse direction based on the positioning information. The displacement synchronization control includes:

[0082] The Hall speed sensor installed on each electric motor is used to collect the walking speed data, and the PID control is used to perform closed-loop speed regulation;

[0083] The laser ranging sensor is used to monitor the displacement data of the two tires at the same time, and the displacement data of the two tires are synchronized in combination with the positioning information to perform displacement control.

[0084] The embodiment of the present invention also provides an electrical control system for a dual-tyre beam moving machine based on the method, referring to Figure 1, including a positioning data acquisition unit, a running synchronization control unit, a leg tilt and stress monitoring unit, a tire pressure detection unit, a wire rope detection unit, a main beam deflection monitoring unit and a remote monitoring unit connected in a CAN bus networking manner, and the positioning data acquisition unit includes a GNSS system and an RTK system.

[0085] The walking synchronization control unit includes a Hall speed sensor, a laser ranging sensor installed at each motor, and a controller for collecting data monitored by the Hall speed sensor and the laser ranging sensor. The walking synchronization control unit realizes displacement synchronization through indirect measurement synchronization control and direct measurement synchronization control.

[0086] Indirect measurement synchronous control includes: each electric motor of the beam handling machine is equipped with a Hall speed sensor to monitor the driving speed. The controller obtains the speed signal of the electric motor and realizes closed-loop regulation of the vehicle speed through proportional-integral-differential (PID) control. This method ensures the stability and controllability of the handling machine during the movement process.

[0087] Direct measurement synchronization control includes: by adopting dual-machine wireless data sharing, the two devices are closely connected to achieve soft synchronization of motor speed measurement. At the same time, the real-time displacement can be monitored with the help of laser ranging sensors to further improve the accuracy of synchronization control.

[0088] The autonomous driving intelligent collision avoidance unit includes an inertial navigation system, cameras and radar sensors.

[0089] The leg inclination and stress monitoring unit includes an angle tilter installed on the leg and a vibrating wire strain gauge installed inside the mold frame, which monitors the inclination of the leg of the beam transporter and the stress changes of the beam transporter mold frame itself in real time during the travel process. By installing an angle tilter on the leg and installing a vibrating wire strain gauge in the steel of the mold frame itself, it is used to determine whether the beam transporter is in a safe state during the travel process, and brake operation is performed through the electrical control system when necessary. In this embodiment, firstly, in terms of leg inclination monitoring, a high-precision angle tilter needs to be accurately installed on each leg to ensure that it is firmly combined with the leg. By correctly connecting the output of the angle tilter to the electrical control system, the real-time monitoring of the leg inclination is achieved. Secondly, for the monitoring of the mold frame stress state, the vibrating wire strain gauge should be accurately installed in the steel of the mold frame. Ensure that the strain gauge has a high degree of fit with the mold frame, and connect its output to the electrical control system. In this way, the changes in the stress state of the mold frame can be accurately sensed during operation. In terms of the configuration of the electrical control system, the data of the leg inclination and the mold frame stress state are processed by corresponding algorithms. Establish appropriate logic to determine whether the beam transporter is in a safe state, including setting safe inclination and stress thresholds. Finally, by setting the response threshold judgment logic of the brake system, the electrical control system can take braking actions in time to ensure the stability of the equipment when it detects abnormal outrigger inclination or formwork stress state. This complete operation process ensures that the outrigger inclination and stress monitoring unit can effectively monitor and respond to potential safety risks in real time during the walking process, thereby maintaining the stability of the beam transporter.

[0090] The tire pressure detection unit includes a tire pressure gauge and a thermometer installed on each tire, and each tire pressure gauge and thermometer is electrically connected to a quantitative repeater for real-time monitoring, display and alarm of the tire pressure of the transporter. The quantitative repeater can be installed and calibrated on site to achieve monitoring, display and alarm of the tire pressure and temperature of 88 tires. In this embodiment, it is first necessary to install a quantitative repeater on each tire. During on-site operation, these quantitative repeaters are accurately calibrated to ensure that the readings of each sensor accurately reflect the tire pressure and temperature of the corresponding tire. This includes adjusting the sensitivity, range and calibration of the sensor. Next, these quantitative repeaters are connected to the tire pressure monitoring device to ensure that the monitoring system can accurately receive, interpret and record the tire pressure and temperature data of each tire. Through the visual interface, the user can intuitively view the status of all tires. In order to ensure the real-time performance of the system, an alarm mechanism is configured so that it can issue an alarm in time when abnormal tire pressure or temperature is detected. During the entire monitoring process, data is stored in real time and historical data is archived so that users can view past tire status and perform trend analysis. This helps to predict possible problems and take corresponding preventive measures. The system can also generate reports that provide a comprehensive assessment of tire performance.

[0091] The wire rope detection unit adopts a wire rope detection device to realize the quantitative and visual functions of wire rope defects. By configuring the wire rope detection device for the equipment, the degree of damage, the exact location of the damage point, and the electromagnetic waveform curve of the wire rope can be detected, and a corresponding report can be generated, so as to realize the quantitative detection of wire rope defects, thereby making the detection results more intuitive and visual. In this embodiment, a dedicated wire rope detection device is first installed on the equipment to ensure that the device can be tightly fixed on the wire rope of the handling machine to ensure the accuracy and reliability of signal acquisition. Next, the wire rope detection device is calibrated to adjust the sensitivity, sampling frequency and signal processing parameters of the detection device to adapt to wire ropes of different diameters and materials. During the calibration process, different defect conditions are simulated to ensure that the detection unit can accurately respond to various defects. Subsequently, the wire rope of the handling machine passes through the detection device one by one. During the detection process, it is ensured that the equipment can capture the electromagnetic waveform curve of the wire rope and accurately measure various parameters, such as the degree of damage and the location of the damage point. In addition, the detection results are monitored in real time to ensure the timeliness of the data and generate a report. Configure the system to automatically generate reports including test results, parameter values, waveform curves, etc. The report can clearly present the status of each wire rope, allowing users to intuitively understand the defects. Finally, set thresholds and alarm mechanisms in the system. When obvious defects are detected in the wire rope, the system can issue an alarm to remind the operator to take timely maintenance or replacement measures to ensure the safety and reliability of the handling machine.

[0092] The main beam deflection monitoring unit adopts a deflection monitoring device to realize static monitoring of the main beam deflection deformation, thereby achieving millimeter-level static deflection monitoring. In this embodiment, a dedicated main beam deflection monitoring device is first configured on the equipment to ensure that the monitoring device can be firmly installed on the main beam to accurately sense the deflection deformation of the main beam. Next, the main beam deflection monitoring unit is initialized and calibrated. The monitoring unit is initialized and started and ensured to operate normally. During the calibration process, the parameters of the monitoring unit are adjusted by applying an external force of known deflection to ensure the accuracy and precision of deflection monitoring. Subsequently, the main beam of the transporter is placed in an actual working environment for real-time deflection monitoring. The main beam deflection monitoring unit collects the deflection data of the main beam in real time according to a preset sampling frequency. A threshold is preset to trigger an alarm or report when the main beam deflection exceeds a predetermined range. The setting of the sampling frequency can take into account the needs of the actual working environment to ensure timely monitoring of the deflection deformation. The configuration system generates a detailed deflection monitoring report. The report should include real-time deflection data of the main beam, historical deflection change trends, event records exceeding the threshold, and other contents. Ensure that the report provides users with clear information on girder deflection monitoring. Finally, set up remote access and notification functions so that operators can monitor girder deflection at any time through cloud platforms or other means. This helps to detect potential problems in a timely manner and improve the safety and reliability of the equipment.

[0093] The remote monitoring unit is used to present the equipment in the form of digital twins on the cloud platform to realize the cloud-edge collaboration of equipment operation monitoring. The specific steps include the configuration of equipment parameter sensors, data transmission and digital twin modeling to ensure that the key operating parameters of the equipment are accurately presented on the cloud platform. In this embodiment, first, for the inclination angle of the main beam and the outrigger, wind speed, working environment humidity, lifting weight, hook lifting height and limit position, hoisting trolley running stroke and limit position, trolley running stroke and limit position, wheel steering angle, hydraulic system pressure and humidity, front hoisting trolley and rear hoisting trolley running distance monitoring, whole machine horizontal posture and outrigger verticality and other parameters, the corresponding sensors are configured one by one to ensure their accuracy and precision. Secondly, an efficient data transmission channel is established to ensure that real-time working data can be safely and stably transmitted to the cloud platform. Next, digital twin modeling is performed, that is, the physical characteristics of the equipment and the data collected by the sensor are combined to establish a real and dynamic equipment model. Finally, through the cloud-edge collaboration method, the key operating parameters of the equipment are accurately presented on the cloud platform. Through this complete operation process, the remote monitoring unit can efficiently monitor and analyze the equipment operating status on the cloud platform, providing strong support for equipment operation and maintenance.

[0094] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0095] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An electrical control method for a dual-tyre beam moving machine based on intelligent redundant switching. Features The process includes: Step 1: Use the GNSS system and the RTK system to simultaneously acquire location data, and use the environmental sensor unit to acquire environmental data; Step 2: Input the location data and environmental data of the GNSS system and / or RTK system as input data into the LSTM-VAE model for potential data decision processing, and select the optimal measurement source data; Step 3: Calculate positioning information using optimal measurement source data; Step 4: The running synchronization control unit performs synchronous control of the straight and / or lateral displacement of the dual tires based on the positioning information.

2. According to claim 1, the electrical control method of the dual-tyre beam moving machine based on intelligent redundant switching, Features: The environmental data described in step 2 includes weather conditions, terrain information and surrounding building information.

3. The electrical control method of the dual-tyre beam moving machine based on intelligent redundant switching according to claim 1, Features: The sensor information data described in step 2 includes the sensor's signal-to-noise ratio, positioning error and signal quality.

4. The electrical control method of the dual-tyre beam moving machine based on intelligent redundant switching according to claim 1, Features: Step 2 specifically includes the following processes: S2.

1. Convert the location data and environmental data of the GNSS system and / or RTK system into an input sequence and input it into the LSTM network to obtain the LSTM hidden state: h t =LSTM(x t ,h t-1 ) Among them, x t is the position data and environmental data of the GNSS system and / or RTK system at time t, h t is the hidden state of LSTM at time t; S2.

2. The encoder is used to map the LSTM hidden state to the mean and variance parameters of the latent space, expressed as: m, s 2 =Encoder(h t ) Among them, μ and σ 2 denote the mean and variance of the latent space respectively; A latent variable z is sampled from the mean and variance of the encoder output, expressed as: z~N(μ,σ 2 ) Where N represents the number of samples; S2.

3. Use the decoder to decode the latent variable z into a reconstruction of the input sequence, expressed as: x t '=Decoder(z) Among them, x t ' is to reconstruct sequence data; S2.

4. Calculate the reconstruction loss and KL divergence loss. The reconstruction loss is the difference between the reconstructed sequence data and the input sequence, and is calculated using the mean square error function: Among them, Reconstruction Loss represents the reconstruction loss; The KL divergence loss is the difference between the distribution of the latent variable and the standard normal distribution and is calculated using the following formula: Among them, KLDivergenceLoss represents the KL divergence loss, and J is the dimension of the latent space; S2.

5. Train through back propagation and gradient descent algorithms to minimize reconstruction loss and KL divergence loss, obtain the optimal solution, output the reconstructed sequence data, and form the best measurement source data.

5. The electrical control method of the dual-tyre beam moving machine based on intelligent redundant switching according to claim 4, Features: The input sequence described in step S2.1 is obtained by dynamic selection processing of the measurement source, The process includes: a1. According to the current environmental factors and sensor information data, the location data and environmental data of the GNSS system and RTK system are scored respectively; a2. Take the location data and environmental data corresponding to the source with the high score as the input sequence.

6. The electrical control method of the dual-tyre beam moving machine based on intelligent redundant switching according to claim 1, Features: The displacement synchronization control described in step 4 includes: The Hall speed sensor installed on each electric motor is used to collect the walking speed data, and the PID control is used to perform closed-loop speed regulation; The laser ranging sensor is used to monitor the displacement data of the two tires at the same time, and the displacement data of the two tires are synchronized in combination with the positioning information to perform displacement control.

7. An electrical control system for a dual-tyre beam moving machine based on the method of claim 1, Features: It includes a positioning data acquisition unit, a running synchronization control unit, an automatic driving intelligent anti-collision unit, a leg tilt and stress monitoring unit, a tire pressure detection unit, a wire rope detection unit, a main beam deflection monitoring unit and a remote monitoring unit connected in a CAN bus networking manner. The positioning data acquisition unit includes a GNSS system and an RTK system.

8. The electrical control system of the dual-tyre beam moving machine according to claim 8, Features: The walking synchronization control unit includes a Hall speed sensor and a laser ranging sensor installed at each motor, and a Hall speed sensor and a laser ranging sensor for collecting information.

9. The electrical control system of the dual-tyre beam moving machine according to claim 8, Features: The leg tilt and stress monitoring unit comprises an angle tilter installed on the leg and a vibrating wire strain gauge installed inside the mold frame.

10. The electrical control system of the dual-tyre beam moving machine according to claim 8, Features: The autonomous driving intelligent collision avoidance unit includes an inertial navigation system, cameras and radar sensors.

Images