A platform travel speed control method and system for aerial work platform

Through data-driven and intelligent optimization technology, combined with gray prediction and genetic algorithm, the optimal vehicle speed optimization model is built, which solves the intelligent and optimization problems of walking speed control of high-altitude operation platforms, improves operator comfort and platform safety, and realizes the Efficient speed control.

CN118778717BActive Publication Date: 2024-11-29MANTALL HEAVY IND
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411270804.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-29
Estimated Expiration
2044-09-11

AI Technical Summary

Technical Problem

The walking speed control method of the existing high-altitude working platform fails to fully consider the working environment, platform status and operator experience, making it difficult to maximize operator comfort while ensuring safety, and it is difficult to achieve intelligent and optimized speed. adjust.

Method used

Using data-driven and intelligent optimization technology, by collecting operation data and driving data from the aerial work platform, using pre-built comfort analysis model and gray prediction model, combined with genetic algorithms, the optimal vehicle speed optimization model is built, and the control instructions are corrected in real time to achieve Optimal speed control.

Benefits of technology

The intelligent and optimized walking speed of the high-altitude operation platform is realized, the operator comfort and platform safety, stability and efficiency are improved, the impact of sample data fluctuations is reduced, and the speed control is provided for quantitative indicators and fast response.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118778717B_ABST
    Figure CN118778717B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of aerial work platforms, and discloses a platform travel speed control method and system for aerial work platforms, the method comprising the following steps: collecting the operation data and driving data of the aerial work platform; predicting the operator's comfort using a comfort analysis model, and solving the optimal vehicle speed with maximizing the operator's comfort as the optimization goal; performing PID calculation based on speed deviation to achieve real-time correction of control instructions; driving the aerial work platform to move according to the corrected control instructions, and displaying the vehicle speed status, working condition status and fault status information in real time. The present invention can effectively realize the intelligence and optimization of the travel speed control of the aerial work platform by making full use of data-driven and intelligent optimization technologies, and takes into account the safety, stability and efficiency of the platform's travel while improving the operator's comfort, and has good application prospects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of aerial work platforms, and in particular to a method and system for controlling the travel speed of an aerial work platform. Background Art

[0002] Aerial work platform is a kind of special operation equipment widely used in the fields of construction, electricity, communication, aerospace, etc. It uses hydraulic, electric and other driving methods to send the work platform and the operator to the aerial work position to achieve tasks such as aerial work, maintenance, cleaning, decoration, etc. Compared with traditional aerial work methods such as scaffolding and hanging basket, aerial work platform has the advantages of high safety, strong flexibility and high working efficiency, and has become the mainstream equipment for modern aerial work. Aerial work platform can be divided into scissor type, curved arm type, straight arm type, mast type and other types according to the characteristics of the platform, and can be divided into self-propelled, vehicle-mounted, trailer type and other types according to the walking mode. Different types of aerial work platforms have their own characteristics in terms of performance, scope of application, operation mode and so on. With the widespread application of aerial work platform and technological progress, its development trend of intelligence, unmanned and networking is becoming increasingly obvious, providing new possibilities for improving the safety and efficiency of operations.

[0003] During the operation of the aerial work platform, the travel speed of the platform is one of the key factors affecting the safety of the operation and the comfort of the operator. If the travel speed is too low, the operation efficiency will be reduced and the construction period will be prolonged; if the travel speed is too high, it may cause safety accidents such as overturning and instability of the platform, and it will also aggravate the vibration of the platform and reduce the comfort of the operator. Therefore, how to adjust the travel speed of the aerial work platform in real time according to the working environment and working conditions, so as to maximize the comfort of the operator while ensuring safety, is an urgent problem to be solved.

[0004] At present, there are mainly the following ways to control the walking speed of aerial work platforms:

[0005] 1) Manual control: The operator manually adjusts the platform speed based on his own experience and perception. This method completely relies on the operator's subjective judgment and lacks objective data support, making it difficult to ensure the optimal speed. The experience levels of different operators vary greatly, and the control effect is difficult to unify. At the same time, the operator is prone to fatigue and mistakes when manually controlling the speed for a long time in an aerial working environment, which poses certain safety hazards.

[0006] 2) Constant speed control: The platform moves at a preset fixed speed. The speed parameters are set at the factory and cannot be adjusted in real time. This method is simple to operate, but it cannot adaptively adjust the speed according to changes in operating conditions. When factors such as platform load, operating height, wind load, slope, etc. change, the fixed travel speed is difficult to meet safety and comfort requirements. Especially in complex operating environments, constant speed control is prone to cause safety and comfort issues.

[0007] 3) Single factor feedback control: The travel speed is adjusted through feedback control according to a single factor such as the platform's load and height. This method has a certain degree of adaptability compared to constant speed control, but it only considers the influence of a single factor and ignores the coupling effect between multiple factors. In actual operations, factors such as load, height, wind load, and slope often affect each other, and single factor feedback control is difficult to achieve optimal speed regulation.

[0008] In summary, the above speed control method does not fully consider multiple factors such as the working environment, platform status and operator experience, and cannot maximize the operator comfort while ensuring safety. The intelligence and optimization of the platform travel speed control needs to be improved.

[0009] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0010] In view of the problems in the related art, the present invention proposes a platform travel speed control method and system for an aerial work platform to overcome the above-mentioned technical problems existing in the existing related art.

[0011] To this end, the specific technical solution adopted by the present invention is as follows:

[0012] According to one aspect of the present invention, a method for controlling the platform travel speed of an aerial work platform is provided, and the method for controlling the platform travel speed of the aerial work platform comprises the following steps:

[0013] S1. Collecting operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed;

[0014] S2. Use the pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal;

[0015] S3, calculating the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and performing PID operation based on the speed deviation to realize real-time correction of the control command;

[0016] S4. Drive the aerial work platform to move according to the corrected control instructions, and display the vehicle speed status, working condition status and fault status information in real time.

[0017] Furthermore, the operation data and driving data of the aerial work platform are collected by various sensors pre-installed on the aerial work platform.

[0018] Furthermore, the method of using the pre-built comfort analysis model in combination with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solving the optimal vehicle speed with maximizing the operator's comfort as the optimization goal includes the following steps:

[0019] S21, constructing a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and performing training;

[0020] S22, using the trained comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort;

[0021] S23. Construct an optimal vehicle speed optimization model, take maximizing operator comfort as the optimization goal, and use safety constraints as optimization conditions to solve the optimal vehicle speed under the current working conditions.

[0022] Furthermore, the construction of a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and the training includes the following steps:

[0023] S211, obtaining operation data, driving data and operator comfort evaluation data of the aerial work platform from a historical database;

[0024] S212. Select key factors that affect the operator's comfort as input variables of the comfort analysis model, select the operator's comfort evaluation as the output variable of the comfort analysis model, construct the comfort analysis model and perform training.

[0025] Furthermore, the key factors affecting the operator's comfort are selected as input variables of the comfort analysis model, the operator's comfort evaluation is used as the output variable of the comfort analysis model, and the comfort analysis model is constructed and trained, including the following steps:

[0026] S2121. Select the load, working height, wind load, slope and vehicle speed of the aerial work platform as input variables of the comfort analysis model, and the operator's comfort evaluation as the output variable of the comfort analysis model to obtain sample data;

[0027] S2122. Using the grey correlation clustering method, select the input variables in the sample data whose correlation is lower than the preset threshold, and use the grey prediction model to correct the sample data;

[0028] S2123. Construct a grey multivariate linear regression model based on the corrected sample data and perform training.

[0029] Furthermore, the method of using the grey correlation clustering method to screen out input variables in the sample data whose correlation is lower than a preset threshold, and using the grey prediction model to correct the sample data includes the following steps:

[0030] S21221. Calculate the grey absolute correlation degree of the sample data to obtain a grey absolute correlation degree matrix;

[0031] S21222. Perform cluster analysis on the grey absolute correlation matrix, traverse all elements in the grey absolute correlation matrix, and form several input variables according to the clustering principle;

[0032] S21223. Calculate the average value or minimum value of the grey absolute correlation degree of each input variable as the correlation degree of the input variable;

[0033] S21224. Selecting the input variable with the lowest correlation from each input variable as a representative according to a preset selection strategy to form a subset of the screened input variables;

[0034] S21225. Use the grey prediction model to correct the screened input variable sequence and output variable sequence.

[0035] Furthermore, the gray prediction model is used to modify the screened input variable sequence, which includes the following steps:

[0036] Based on each input variable sequence after screening, a grey prediction model is established; each input variable sequence is accumulated once to obtain an accumulated generation sequence based on the input variable, and the accumulated generation sequence based on the input variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the input variable;

[0037] The first-order linear differential equation between the cumulative generation sequence based on the input variables and the adjacent mean sequence based on the input variables is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the input variables is obtained;

[0038] The grey prediction model based on the input variable sequence is obtained by solving the whitening equation. The grey prediction model based on the input variable sequence is used to predict the cumulative generation sequence based on the input variables to obtain the predicted cumulative generation sequence of the input variables, and the predicted cumulative generation sequence of the input variables is restored to obtain the corrected value sequence of the input variables.

[0039] Furthermore, the gray prediction model is used to correct the output variable sequence after screening, including the following steps:

[0040] Based on each output variable sequence after screening, a grey prediction model is established; each output variable sequence is accumulated once to obtain an accumulated generation sequence based on the output variable, and the accumulated generation sequence based on the output variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the output variable;

[0041] The first-order linear differential equation between the cumulative generation sequence based on the output variable and the adjacent mean sequence based on the output variable is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the output variable is obtained;

[0042] Solving the whitening equation of the output variable obtains the grey prediction model based on the output variable sequence, using the grey prediction model based on the output variable sequence to predict the cumulative generation sequence based on the output variable to obtain the predicted cumulative generation sequence of the output variable, and restoring the predicted cumulative generation sequence of the output variable to obtain the corrected value sequence of the output variable.

[0043] Furthermore, the optimal vehicle speed optimization model is constructed, with the operator comfort maximization as the optimization goal and the safety constraint as the optimization condition, and solving the optimal vehicle speed under the current working condition includes the following steps:

[0044] S231, defining the speed of the aerial work platform as an optimization variable, and determining a value range of the speed;

[0045] S232, constructing an objective function based on the predicted relationship between the operator's comfort and the vehicle speed, introducing a vehicle speed constraint condition based on the safety requirements of the aerial work platform, and converting the dynamic stability constraint, the braking distance constraint, and the obstacle avoidance constraint into an inequality constraint for the vehicle speed;

[0046] S233, integrating the objective function and the constraint conditions to construct an optimal vehicle speed optimization model, and using a genetic algorithm to combine the current operating data and driving data to solve the optimal vehicle speed.

[0047] According to another aspect of the present invention, a platform travel speed control system for an aerial work platform is provided, the platform travel speed control system for an aerial work platform comprises: a data acquisition module, an optimal vehicle speed calculation module, a control instruction correction module and a drive control module, and the data acquisition module is connected to the optimal vehicle speed calculation module, the control instruction correction module and the drive control module in sequence;

[0048] The data acquisition module is used to collect the operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed;

[0049] The optimal vehicle speed calculation module is used to predict the operator's comfort by using a pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform, and to solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal;

[0050] The control instruction correction module is used to calculate the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and perform PID operation based on the speed deviation to realize real-time correction of the control instruction;

[0051] The drive control module is used to drive the aerial work platform to move according to the corrected control instructions, and to display the vehicle speed status, working condition status and fault status information in real time.

[0052] The beneficial effects of the present invention are:

[0053] 1) The present invention can effectively realize the intelligentization and optimization of the walking speed control of the aerial work platform by making full use of data-driven and intelligent optimization technologies, while improving the comfort of the operator and taking into account the safety, stability and efficiency of the platform's walking, and has good application prospects.

[0054] 2) The present invention uses sensors to collect operation data and driving data in real time, comprehensively perceives the platform status and operating environment, and provides a reliable basis for speed optimization. It also introduces a comfort analysis model based on historical data and uses grey correlation clustering and grey prediction to correct sample data, thereby effectively reducing the impact of sample data fluctuations. The corrected data can then be used to construct a more accurate grey multivariate linear regression model to predict the operator's comfort, providing quantitative indicators for speed optimization.

[0055] 3) The present invention takes maximizing operator comfort as the optimization goal, comprehensively considers safety constraints to construct an optimal vehicle speed optimization model, and uses genetic algorithm to solve the dynamic optimal vehicle speed to achieve optimization and adaptation of speed control. Therefore, PID operation can be performed according to the speed deviation to achieve real-time correction of control instructions, thereby ensuring the stability and rapid responsiveness of speed control. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention 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 the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 The present invention is a flowchart of a method for controlling the travel speed of an aerial work platform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0059] According to an embodiment of the present invention, a method and system for controlling the travel speed of an aerial work platform are provided.

[0060] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a platform travel speed control method for an aerial work platform is provided, and the platform travel speed control method for an aerial work platform comprises the following steps:

[0061] S1. Collecting operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed;

[0062] Specifically include:

[0063] 1) Collection of load data: Install a weighing sensor, such as a pressure sensor or a weighing meter, on the working platform of the aerial work platform; measure the load weight on the working platform in real time, including the total weight of the operator, tools, materials, etc.; associate the load weight data with the timestamp to form a load data time series.

[0064] 2) Collection of working height data: Use height sensors, such as laser rangefinders, ultrasonic rangefinders or GPS positioning systems, to measure the working height of aerial work platforms; record the height data of the working platform in real time to form a time series of working height data; associate the collected working height data with timestamps;

[0065] 3) Collection of wind load data: Install wind speed and direction sensors, such as ultrasonic anemometers or mechanical anemometers, on the aerial work platform; measure the wind speed and direction data in the working environment in real time; associate the wind speed and direction data with timestamps to form a wind load data time series.

[0066] 4) Slope data collection: Use inclination sensors, such as accelerometers or electronic levels, to measure the pitch and roll angles of the aerial work platform; record the inclination data of the aerial work platform in real time to form a slope data time series; associate the collected slope data with the timestamp.

[0067] 5) Collection of real-time vehicle speed data: Use a vehicle speed sensor, such as an electromagnetic vehicle speed sensor or a GPS speed measurement system, to measure the travel speed of the aerial work platform in real time; associate the vehicle speed data with the timestamp to form a real-time vehicle speed data time series.

[0068] 6) Data synchronization and preprocessing: synchronize load data, operating height data, wind load data, slope data and real-time vehicle speed data. Align each data time series according to the timestamp to ensure the consistency of different types of data in the time dimension. Preprocess the collected data, such as denoising, outlier processing, data normalization, etc., to improve data quality.

[0069] S2. Use the pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal;

[0070] The method of using a pre-built comfort analysis model in combination with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solving the optimal vehicle speed with maximizing the operator's comfort as the optimization goal includes the following steps:

[0071] S21, constructing a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and performing training;

[0072] Specifically, the construction of a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and the training includes the following steps:

[0073] S211, obtaining operation data, driving data and operator comfort evaluation data of the aerial work platform from a historical database;

[0074] S212, selecting key factors affecting operator comfort as input variables of the comfort analysis model, the operator's comfort evaluation as the output variable of the comfort analysis model, constructing the comfort analysis model and training it; specifically comprising the following steps:

[0075] S2121. Select the load, working height, wind load, slope and vehicle speed of the aerial work platform as input variables of the comfort analysis model, and the operator's comfort evaluation as the output variable of the comfort analysis model to obtain sample data;

[0076] The operator's comfort evaluation includes: subjective scoring (that is, the operator makes subjective scoring of each dimension based on his actual experience under different working conditions) and physiological parameter measurement (that is, evaluating the operator's physiological comfort level under different working conditions based on the changes in physiological parameters). The results of subjective scoring and physiological parameter measurement can be comprehensively analyzed, and the comprehensive comfort evaluation of the operator under different working conditions can be obtained through weighted average, principal component analysis and other methods. The comprehensive evaluation can quantify the operator's overall comfort level as the output variable of the comfort analysis model.

[0077] S2122, using the grey correlation clustering method to screen out input variables in the sample data whose correlation is lower than a preset threshold, and using the grey prediction model to correct the sample data; specifically comprising the following steps:

[0078] S21221. Calculate the grey absolute correlation degree of the sample data to obtain a grey absolute correlation degree matrix; specifically including:

[0079] 1) Preprocess the sample data:

[0080] Assume that there are m samples, each of which contains n input variables and 1 output variable. Organize the sample data into an m×(n+1) matrix X, where the first n columns are input variables and the last column is the output variable. Normalize the matrix X and scale the value range of each variable to the interval [0,1] to eliminate the dimension effect.

[0081] 2) Calculate the grey absolute correlation between the input variable and the output variable: For the i-th sample (i=1,2,...,m), calculate its input variable x i With the output variable y i The gray absolute correlation between them; calculate the gray absolute correlation of all samples and get an m-dimensional vector γ, where γ i Represents the grey absolute correlation degree of the i-th sample.

[0082] 3) Calculate the grey absolute correlation between input variables: For any two input variables x j and x k (j,k=1,2,...,n;j≠k), calculate the grey absolute correlation between them; calculate the grey absolute correlation between all input variables, and get an n×n symmetric matrix R, where R jk Represents the grey absolute correlation between the j-th and k-th input variables.

[0083] 4) Constructing the grey absolute correlation matrix: Combining the m-dimensional vector γ obtained in 2) and the n×n matrix R obtained in 3) to construct a (m+n)×(n+1) grey absolute correlation matrix G. The grey absolute correlation matrix G comprehensively reflects the correlation strength between the input variables and the output variables, as well as the correlation between the input variables.

[0084] S21222. Perform cluster analysis on the grey absolute correlation matrix, traverse all elements in the grey absolute correlation matrix, and form several input variables according to the clustering principle; specifically including:

[0085] 1) Determine the clustering principle: divide the input variables into different clusters according to the size of the grey absolute correlation degree; the clustering principle can be determined based on the threshold of the grey absolute correlation degree or the clustering algorithm (such as hierarchical clustering, k-means clustering, etc.); for example, a grey absolute correlation degree threshold θ can be set, and when the grey absolute correlation degree between two input variables is greater than or equal to θ, they are classified into the same category.

[0086] 2) Traverse the grey absolute correlation matrix: extract the grey absolute correlation submatrix R between the input variables from the grey absolute correlation matrix G; traverse all elements R in the matrix R jk Compare them with the size of the clustering threshold θ; if R jk ≥θ, then the input variable x j and x k into the same category, otherwise, they are divided into different categories.

[0087] 3) Forming input variable combinations: According to the clustering results of 2), the input variables are divided into several non-overlapping subsets; each subset contains one or more input variables, which have a high correlation in the grey absolute correlation matrix; the input variable combinations formed can be expressed as C1, C2, ..., C q , where q is the number of categories after clustering.

[0088] S21223. Calculate the average value or minimum value of the grey absolute correlation degree of each input variable as the correlation degree of the input variable;

[0089] S21224. Selecting the input variable with the lowest correlation from each input variable as a representative according to a preset selection strategy to form a subset of the screened input variables; specifically including:

[0090] 1) Set the selection strategy: determine the strategy for selecting representative variables from each input variable combination (including selecting the variable with the lowest average correlation with other variables as the representative, selecting the variable with the lowest maximum correlation with other variables as the representative, and selecting the variable with the highest correlation with the output variable as the representative);

[0091] 2) Calculate the correlation index of each variable: For each input variable combination C i (i=1,2,...,q), calculate the correlation index of each variable in the combination; if the selection strategy is average correlation, then calculate the average value of the gray absolute correlation between each variable and other variables; if the selection strategy is maximum correlation, then calculate the maximum value of the gray absolute correlation between each variable and other variables; if the selection strategy is correlation with the output variable, then directly use the gray absolute correlation between each variable and the output variable.

[0092] 3) Select the variable with the lowest correlation as the representative: For each input variable combination C i According to the correlation index calculated in 2), select the variable with the lowest correlation as the representative; record the selected representative variable and add it to the screened input variable subset S; repeat the above process until representative variables are selected for all input variable combinations.

[0093] 4) Forming a filtered input variable subset: All representative variables selected in 3) are combined into a filtered input variable subset S; the subset S contains the variables with the lowest correlation in each input variable combination, and these variables can represent the characteristics of their respective combinations to a certain extent; the filtered input variable subset S will be used as the input feature of the comfort analysis model for subsequent model training and prediction.

[0094] S21225. Use the grey prediction model to correct the screened input variable sequence and output variable sequence.

[0095] The use of the grey prediction model to correct the screened input variable sequence includes the following steps:

[0096] Based on each input variable sequence after screening, a grey prediction model is established; each input variable sequence is accumulated once to obtain an accumulated generation sequence based on the input variable, and the accumulated generation sequence based on the input variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the input variable;

[0097] The first-order linear differential equation between the cumulative generation sequence based on the input variables and the adjacent mean sequence based on the input variables is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the input variables is obtained;

[0098] Solve the whitening equation to obtain the grey prediction model based on the input variable sequence, use the grey prediction model based on the input variable sequence to predict the cumulative generation sequence based on the input variables to obtain the predicted cumulative generation sequence of the input variables, and restore the predicted cumulative generation sequence of the input variables to obtain the corrected value sequence of the input variables (the restoration process is: the corrected value of each period is equal to the difference between the predicted cumulative value of the period and the predicted cumulative value of the previous period. The restored sequence is the corrected value sequence of the input variables, which reflects the new sequence of the original input variable sequence after the model correction).

[0099] The use of the grey prediction model to correct the screened output variable sequence includes the following steps:

[0100] Based on each output variable sequence after screening, a grey prediction model is established; each output variable sequence is accumulated once to obtain an accumulated generation sequence based on the output variable, and the accumulated generation sequence based on the output variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the output variable;

[0101] The first-order linear differential equation between the cumulative generation sequence based on the output variable and the adjacent mean sequence based on the output variable is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the output variable is obtained;

[0102] Solving the whitening equation of the output variable obtains the grey prediction model based on the output variable sequence, using the grey prediction model based on the output variable sequence to predict the cumulative generation sequence based on the output variable to obtain the predicted cumulative generation sequence of the output variable, and restoring the predicted cumulative generation sequence of the output variable to obtain the corrected value sequence of the output variable.

[0103] In order to better understand the steps of using the grey prediction model to correct the screened input variable sequence and output variable sequence, the following example is given as follows:

[0104] 1) For each input variable sequence x after screening i (i=1,2,…,k) and the output variable sequence y, and establish the GM(1,1) model respectively.

[0105] 2) For the input variable sequence x i (i=1,2,…,k), perform an accumulation generation, and obtain the accumulation generation sequence x based on the input variablei (1) .

[0106] 3) For x i (1) Generate the nearest mean and obtain the nearest mean sequence z based on the input variable i (1) .

[0107] 4) Create the original input variable sequence x i (1) and z i (1) A first-order linear differential equation:

[0108] dx i (1) / dt+ax i (1) = b;

[0109] Among them, a is the development coefficient, b is the grey action, and t is the time variable.

[0110] 5) Estimate the parameters a and b using the least squares method to obtain the whitening equation of the GM(1,1) model;

[0111] 6) Solve the whitening equation and get the prediction model:

[0112] x i (1) (t) = [x i (1) (1)-b / a] *e [-a(t-1)] +b / a;

[0113] Among them, x i (1) (t) represents x i (1) The predicted value at time t, x i (1) (1) represents x i (1) The initial value of , e represents the base of natural logarithm;

[0114] 7) For x i (1) (t) is restored to obtain the input variable sequence x i Correction value x i '(t):

[0115] x i '(t) = x i (1) (t)- x i (1) (t-1), t=2,3,…,m;

[0116] 8) Repeat 2) to 7) for the output variable y to obtain the corrected value y' of the output variable sequence.

[0117] 9) The corrected values ​​of all variables are combined into a new sample matrix X'.

[0118] S2123. Construct a grey multivariate linear regression model based on the corrected sample data and perform training.

[0119] S22, using the trained comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort;

[0120] Specifically, the operation data and driving data collected in S1 are first input into the trained grey multivariate linear regression model. Then, the model predicts the operator's comfort index in real time based on the input independent variable data. Finally, the predicted comfort index is compared with the comfort level threshold to determine the current comfort level. If the comfort index exceeds the threshold range, an alarm is triggered or corresponding control measures are taken.

[0121] Comfort levels include:

[0122] Very comfortable: The comfort index is within the optimal range, indicating that the operator is in the most comfortable state. The threshold of the comfort index can be set between 0.8 and 1.0. At this level, all parameters of the aerial work platform are within the optimal working range, and the operator feels the most pleasant physically and mentally.

[0123] Comfort: The comfort index is within an ideal range, indicating that the operator is in a relatively comfortable state. The threshold of the comfort index can be set between 0.6 and 0.8. At this level, the various parameters of the aerial work platform slightly deviate from the optimal working range, but are still within an acceptable range, and the operator feels relatively relaxed physically and mentally.

[0124] General: The comfort index is in the medium range, indicating that the operator is in a generally comfortable state. The threshold of the comfort index can be set between 0.4 and 0.6. At this level, some parameters of the aerial work platform may deviate from the optimal working range, and the operator may feel slight discomfort or fatigue.

[0125] Discomfort: The comfort index is in the lower range, indicating that the operator is in an uncomfortable state. The threshold of the comfort index can be set between 0.2 and 0.4. At this level, multiple parameters of the aerial work platform are significantly deviated from the optimal working range, and the operator may feel obvious discomfort, fatigue or discomfort.

[0126] Very uncomfortable: The comfort index is in the lowest range, indicating that the operator is in an extremely uncomfortable state. The threshold of the comfort index can be set between 0.0 and 0.2. At this level, the various parameters of the aerial work platform seriously deviate from the optimal working range, and the operator's physical and mental feelings are extremely uncomfortable, which may affect the safety and efficiency of the operation.

[0127] When the predicted comfort index exceeds the set threshold range, the system can automatically trigger an alarm to remind the operator to pay attention to the change in comfort. At the same time, corresponding control measures can be taken, such as adjusting the speed and balance state of the aerial work platform, to ensure the operator's comfort and work safety.

[0128] S23. Construct an optimal vehicle speed optimization model, take maximizing operator comfort as the optimization goal, and use safety constraints as optimization conditions to solve the optimal vehicle speed under the current working conditions.

[0129] Specifically, the optimal vehicle speed optimization model is constructed, with the operator comfort maximization as the optimization goal and the safety constraint as the optimization condition, and solving the optimal vehicle speed under the current working condition includes the following steps:

[0130] S231, defining the speed of the aerial work platform as an optimization variable, and determining a value range of the speed, where the value range is determined by the performance and safety requirements of the aerial work platform;

[0131] S232. Construct an objective function max f(v) based on the predicted relationship between the operator's comfort and the vehicle speed, where f(v) represents the function of the operator's comfort with respect to the vehicle speed v. The specific form of f(v) can be fitted and constructed based on the results of the comfort prediction model, such as a linear function, a quadratic function, an exponential function, etc.; introduce vehicle speed constraints based on the safety requirements of the aerial work platform, and transform the dynamic stability constraints, the braking distance constraints, and the obstacle avoidance constraints into inequality constraints for the vehicle speed; that is, g1(v)= vv max ≤0 (dynamic stability constraint), v max represents the upper speed limit under the dynamic stability constraint; g2(v)= v 2 / (2·a max ) -d safe ≤0 (braking distance constraint), a max Indicates the maximum braking acceleration, d safe Indicates safe braking distance; g3(v)=v·t react + v 2 / (2·a max ) - d obstacle ≤0 (obstacle avoidance constraint), t react represents the obstacle avoidance reaction time, d obstacle Indicates the obstacle distance;

[0132] Dynamic stability constraint: Simplify the aerial work platform into a dynamic model, establish the relationship between vehicle speed and dynamic parameters (such as overturning moment, lateral acceleration, etc.), and ensure that the dynamic parameters meet the stability requirements at the vehicle speed;

[0133] Braking distance constraint: Calculate the braking distance based on vehicle speed and braking performance parameters to ensure that the vehicle can stop in time in an emergency to avoid collision;

[0134] Obstacle avoidance constraints: Consider the impact of vehicle speed on obstacle perception and avoidance capabilities, ensuring that the aerial work platform can detect and avoid obstacles in a timely manner at a given vehicle speed;

[0135] S233, integrating the objective function and the constraint conditions to construct an optimal vehicle speed optimization model, and using a genetic algorithm to combine the current operating data and driving data to solve the optimal vehicle speed.

[0136] The use of genetic algorithms combined with current operating data and driving data to solve the optimal vehicle speed specifically includes:

[0137] 1) Initialize genetic algorithm parameters

[0138] Population size: N represents the number of individuals in each generation.

[0139] Crossover probability: Pc represents the probability of two individuals performing a crossover operation.

[0140] Mutation probability: Pm represents the probability of an individual mutation.

[0141] Maximum number of iterations: T represents the termination condition of the genetic algorithm.

[0142] Individual coding method: real number coding, each individual is represented by a real number, that is, the value of the vehicle speed v.

[0143] 2) Randomly generate the initial population

[0144] Within the scope of the constraints, N vehicle speed values ​​are randomly generated as individuals of the initial population.

[0145] For each individual, calculate its fitness value, that is, the value of the objective function f(v).

[0146] 3) Perform genetic algorithm iteration

[0147] Selection operation: Based on the fitness value of the individual, N individuals are selected from the current population as parents using methods such as roulette or tournament selection.

[0148] Crossover operation: Perform pairwise crossover on parent individuals with probability Pc to generate new offspring individuals.

[0149] Mutation operation: mutate the offspring individuals with probability Pm and introduce new genes.

[0150] Constraint processing: Perform constraint checks on offspring individuals after crossover and mutation, and repair or remove individuals that violate the constraints.

[0151] Fitness evaluation: Calculate the fitness value of the offspring individual, that is, the value of the objective function f(v).

[0152] Population update: merge the parent and child individuals, and select the N best individuals as the new population based on the fitness value.

[0153] 4) Output the optimal solution

[0154] In the final population, the individual with the highest fitness value is selected as the optimal solution, that is, the optimal vehicle speed v*.

[0155] Apply the optimal vehicle speed v* to the speed control system of the aerial work platform to achieve a balance between comfort and safety.

[0156] S3, calculating the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and performing PID operation based on the speed deviation to realize real-time correction of the control command;

[0157] The method of calculating the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and performing PID operation based on the speed deviation to realize real-time correction of the control instruction includes the following steps:

[0158] First, the current speed of the aerial work platform needs to be collected in real time, denoted as v actual , obtain the optimal vehicle speed in the current state from the optimal vehicle speed optimization model, denoted as v optimal , then calculate the velocity deviation: e(t) = v optimal -v actual Next, the PID controller is designed, and in order to implement the PID controller in the digital control system, it needs to be discretized; the controller output is calculated according to the discretized PID controller recursive formula. The control command of the aerial work platform is corrected. The corrected control command is sent to the actuator of the aerial work platform.

[0159] S4. Drive the aerial work platform to move according to the corrected control instructions, and display the vehicle speed status, working condition status and fault status information in real time.

[0160] According to another embodiment of the present invention, a platform travel speed control system for an aerial work platform is provided, the platform travel speed control system for an aerial work platform comprises: a data acquisition module, an optimal vehicle speed calculation module, a control instruction correction module and a drive control module, and the data acquisition module is connected to the optimal vehicle speed calculation module, the control instruction correction module and the drive control module in sequence;

[0161] The data acquisition module is used to collect the operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed;

[0162] The optimal vehicle speed calculation module is used to predict the operator's comfort by using a pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform, and to solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal;

[0163] The control instruction correction module is used to calculate the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and perform PID operation based on the speed deviation to realize real-time correction of the control instruction;

[0164] The drive control module is used to drive the aerial work platform to move according to the corrected control instructions, and to display the vehicle speed status, working condition status and fault status information in real time.

[0165] In summary, with the help of the above technical scheme of the present invention, the present invention can effectively realize the intelligence and optimization of the walking speed control of the aerial work platform by making full use of data-driven and intelligent optimization technology, while improving the comfort of the operator, taking into account the safety, stability and efficiency of the platform walking, and has good application prospects. In addition, the present invention collects operation data and driving data in real time through sensors, comprehensively perceives the platform status and working environment, and provides a reliable basis for speed optimization. By introducing a comfort analysis model based on historical data, gray correlation clustering and gray prediction are used to correct sample data, so as to effectively reduce the influence of sample data fluctuations, and then the corrected data can be used to construct a gray multivariate linear regression model to predict the comfort of the operator, providing quantitative indicators for speed optimization. In addition, the present invention takes maximizing the comfort of the operator as the optimization goal, comprehensively considers the safety constraints to construct the optimal vehicle speed optimization model, uses genetic algorithm to solve and obtain the dynamic optimal vehicle speed, realizes the optimization and adaptation of speed control, and can perform PID operation according to the speed deviation to realize real-time correction of the control instruction, and ensure the stability and rapid responsiveness of the speed control.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling the platform travel speed of an aerial work platform, characterized in that: The platform travel speed control method for an aerial work platform comprises the following steps: S1. Collecting operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed; S2. Use the pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal; S3, calculating the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and performing PID operation based on the speed deviation to realize real-time correction of the control command; S4, driving the aerial work platform to move according to the corrected control instructions, and displaying the vehicle speed status, working condition status and fault status information in real time; The method of using the pre-built comfort analysis model in combination with the operation data and driving data of the aerial work platform to predict the operator's comfort, and solving the optimal vehicle speed with maximizing the operator's comfort as the optimization goal includes the following steps: S21, constructing a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and performing training; The method of constructing a comfort analysis model based on the operation data, driving data and operator comfort evaluation data of the aerial work platform in the historical database and training the model comprises the following steps: S211, obtaining operation data, driving data and operator comfort evaluation data of the aerial work platform from a historical database; S212, selecting key factors affecting operator comfort as input variables of the comfort analysis model, and the operator's comfort evaluation as output variables of the comfort analysis model, constructing the comfort analysis model and performing training; The key factors affecting the operator's comfort are selected as input variables of the comfort analysis model, the operator's comfort evaluation is used as the output variable of the comfort analysis model, and the comfort analysis model is constructed and trained, including the following steps: S2121. Select the load, working height, wind load, slope and vehicle speed of the aerial work platform as input variables of the comfort analysis model, and the operator's comfort evaluation as the output variable of the comfort analysis model to obtain sample data; S2122. Using the grey correlation clustering method, select the input variables in the sample data whose correlation is lower than the preset threshold, and use the grey prediction model to correct the sample data; The method of using the grey correlation clustering method to screen out input variables whose correlation is lower than a preset threshold in the sample data, and using the grey prediction model to correct the sample data includes the following steps: S21221. Calculate the grey absolute correlation degree of the sample data to obtain a grey absolute correlation degree matrix; S21222. Perform cluster analysis on the grey absolute correlation matrix, traverse all elements in the grey absolute correlation matrix, and form several input variables according to the clustering principle; S21223. Calculate the average value or minimum value of the grey absolute correlation degree of each input variable as the correlation degree of the input variable; S21224. Selecting the input variable with the lowest correlation from each input variable as a representative according to a preset selection strategy to form a subset of the screened input variables; S21225. Modify the screened input variable sequence and output variable sequence using a grey prediction model; The use of the grey prediction model to correct the screened input variable sequence includes the following steps: Based on each input variable sequence after screening, a grey prediction model is established; each input variable sequence is accumulated once to obtain an accumulated generation sequence based on the input variable, and the accumulated generation sequence based on the input variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the input variable; The first-order linear differential equation between the cumulative generation sequence based on the input variables and the adjacent mean sequence based on the input variables is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the input variables is obtained; Solving the whitening equation to obtain a grey prediction model based on an input variable sequence, using the grey prediction model based on the input variable sequence to predict the cumulative generation sequence based on the input variables to obtain a predicted cumulative generation sequence of the input variables, and restoring the predicted cumulative generation sequence of the input variables to obtain a corrected value sequence of the input variables; The use of the grey prediction model to correct the screened output variable sequence includes the following steps: Based on each output variable sequence after screening, a grey prediction model is established; each output variable sequence is accumulated once to obtain an accumulated generation sequence based on the output variable, and the accumulated generation sequence based on the output variable is generated by the adjacent mean to obtain an adjacent mean sequence based on the output variable; The first-order linear differential equation between the cumulative generation sequence based on the output variable and the adjacent mean sequence based on the output variable is established, and the development coefficient and grey action in the first-order linear differential equation are estimated by the least square method, and the whitening equation of the grey prediction model based on the output variable is obtained; Solving the whitening equation of the output variable to obtain a grey prediction model based on the output variable sequence, using the grey prediction model based on the output variable sequence to predict the cumulative generation sequence based on the output variable to obtain the predicted cumulative generation sequence of the output variable, and restoring the predicted cumulative generation sequence of the output variable to obtain the corrected value sequence of the output variable; S2123. Construct a grey multivariate linear regression model according to the corrected sample data and perform training; S22, using the trained comfort analysis model combined with the operation data and driving data of the aerial work platform to predict the operator's comfort; S23. Construct an optimal vehicle speed optimization model, take maximizing operator comfort as the optimization goal, and use safety constraints as optimization conditions to solve the optimal vehicle speed under the current working conditions.

2. A method for controlling the travel speed of an aerial work platform according to claim 1, characterized in that: The operation data and driving data of the aerial work platform are collected through various sensors pre-installed on the aerial work platform.

3. A method for controlling the platform travel speed of an aerial work platform according to claim 1, characterized in that: The optimal vehicle speed optimization model is constructed, with the operator comfort maximization as the optimization goal and the safety constraint as the optimization condition, and solving the optimal vehicle speed under the current working conditions includes the following steps: S231, defining the speed of the aerial work platform as an optimization variable, and determining a value range of the speed; S232, constructing an objective function based on the predicted relationship between the operator's comfort and the vehicle speed, introducing a vehicle speed constraint condition based on the safety requirements of the aerial work platform, and converting the dynamic stability constraint, the braking distance constraint, and the obstacle avoidance constraint into an inequality constraint for the vehicle speed; S233, integrating the objective function and the constraint conditions to construct an optimal vehicle speed optimization model, and using a genetic algorithm to combine the current operating data and driving data to solve the optimal vehicle speed.

4. A platform travel speed control system for an aerial work platform, used to implement the steps of the platform travel speed control method for an aerial work platform as described in any one of claims 1 to 3, characterized in that: The platform travel speed control system for the aerial work platform comprises: a data acquisition module, an optimal vehicle speed calculation module, a control instruction correction module and a drive control module, and the data acquisition module is connected with the optimal vehicle speed calculation module, the control instruction correction module and the drive control module in sequence; The data acquisition module is used to collect the operation data and driving data of the aerial work platform, wherein the operation data includes load data and operation height data, and the driving data includes wind load data, slope data and real-time vehicle speed; The optimal vehicle speed calculation module is used to predict the operator's comfort by using a pre-built comfort analysis model combined with the operation data and driving data of the aerial work platform, and to solve the optimal vehicle speed with maximizing the operator's comfort as the optimization goal; The control instruction correction module is used to calculate the speed deviation of the aerial work platform according to the collected real-time vehicle speed and the optimal vehicle speed, and perform PID operation based on the speed deviation to realize real-time correction of the control instruction; The drive control module is used to drive the aerial work platform to move according to the corrected control instructions, and to display the vehicle speed status, working condition status and fault status information in real time.

Citation Information

Patent Citations

  • Process industry system prediction module based on crossed relevance time-lag gray correlation analysis

    CN109884892A

  • Method for predicting train comfort degradation caused by strong wind railway wind

    CN112149230A