Data-driven aerial work platform four-wheel multi-objective optimization control method

Abstract

The invention discloses a data-driven aerial work platform four-wheel multi-objective optimization control method, which comprises the following steps of: firstly, constructing a four-wheel hydraulic drive mechanism model of an aerial work platform, performing mechanism model time-varying parameter identification by combining a pulse neural network, performing state prediction by utilizing a parameter-estimated accurate hydraulic model, and performing state prediction by utilizing a parameter-estimated accurate hydraulic model; a multi-objective optimization function containing four-wheel synchronous angle deviation and hydraulic control signals is constructed, a decomposition-based multi-objective optimization algorithm is used for solving, accurate control over four-wheel synchronous steering is achieved, and the four-wheel synchronous steering angle deviation is reduced.

Description

Technical Field

[0001] This invention relates to the field of aerial work platform control, and more particularly to a data-driven four-wheel multi-objective optimization control method for aerial work platforms. Background Technology

[0002] Aerial work platforms, as a widely used type of engineering machinery, can quickly and accurately transport personnel, equipment, and materials to designated heights. They are highly adaptable and efficient, capable of handling complex, ever-changing, and unpredictable construction environments, effectively replacing traditional scaffolding methods. Their safety and reliability are also far superior to other engineering machinery products. With rising labor costs, the demand for flexible and portable aerial work platforms is rapidly increasing, and they are playing an increasingly important role. Therefore, ensuring the overall improvement of aerial work platform performance is of great significance to the industry's development. An aerial work platform consists of a work platform, boom, turntable, and chassis. The chassis typically has an automatic retraction / expansion function to maximize vehicle stability. When traveling at high speed, the axle retracts inward to ensure rapid movement; when the vehicle is working, the axle extends outward to ensure overall stability. The axle extension cylinders are located inside the axle, not in contact with the outside, and are unaffected by external construction conditions, preventing corrosion even in a long-term expanded state. All four wheels are equipped with travel reducers and steering cylinders, allowing simultaneous steering and drive of all four wheels. However, during the operation of aerial work platforms, the operational effectiveness is limited by the precision of the execution system and the influence of complex environments. Often, the aerial work platform can only achieve the expected four-wheel synchronous steering. When the tires are not aligned, it can lead to problems such as heavy steering and excessive tire wear. During travel, the aerial work platform may also deviate from the set direction, thus affecting efficiency or the production process. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies and achieve precise synchronous steering of the four wheels of an aerial work platform. This invention proposes a data-driven multi-objective optimization control method for the four wheels of an aerial work platform. By combining the hydraulic characteristics of four-wheel steering and utilizing pulse neural networks and multi-objective optimization control methods, this invention achieves precise control of the hydraulic system of the four wheels of the aerial work platform, ultimately minimizing four-wheel steering deviation, extending vehicle equipment lifespan, improving production efficiency, and enhancing brand value for OEMs.

[0004] The objective of this invention is achieved through the following technical solution: a data-driven four-wheel multi-objective optimization control method for aerial work platforms, the method comprising:

[0005] A four-wheel dynamics model of the chassis of an aerial work platform is constructed. A spiking neural network is used to estimate the time-varying nonlinear parameters in the model, and the four-wheel dynamics model is used as a state prediction model to obtain the state prediction value at the next moment.

[0006] A hydraulic control sequence is constructed with the objective function of minimizing the deviation between the predicted value and the set steering angle of the wheels of the hydraulically driven aerial work platform chassis. A multi-objective optimization algorithm based on a combined decomposition strategy is used to decompose the four-wheel optimization problem into single-wheel optimization problems. Multiple single-wheel optimization problems are optimized simultaneously to obtain the optimal four-wheel hydraulic control sequence.

[0007] Furthermore, the four-wheel dynamics model of the aerial work platform chassis is specifically a hydraulic system: ;

[0008] ;

[0009] ;

[0010] in For load traffic, For flow gain, The flow pressure coefficient, For valve core displacement, For load pressure, For cylinder displacement, The effective area of ​​the hydraulic cylinder piston. It is the total leakage coefficient of the hydraulic cylinder. For the volume of the hydraulic cylinder, This refers to the total mass of the piston and its load, factored onto the piston. For effective bulk modulus, This is the viscous damping coefficient of the piston and load. For the load spring stiffness, This refers to any external load force acting on the piston.

[0011] Furthermore, the time-varying nonlinear parameter includes flow gain. Flow pressure coefficient and the total leakage coefficient of the hydraulic cylinder .

[0012] Furthermore, the membrane potential equation of the spiking neural network is expressed as follows:

[0013]

[0014] in The membrane time constant is This is the resting potential. For membrane resistance, For synaptic current, when the membrane potential Exceeding the threshold Afterward, the neuron sends out a pulse, and the membrane potential resets to the resting potential. The model parameters of the spiking neural network are dynamically adjusted using a time-triggered mechanism.

[0015] Furthermore, the objective function minimizes the predicted state of the four-wheel rotation angle. The steering angle is set with the wheels of the hydraulically driven aerial work platform chassis. The deviation between them, while taking into account the hydraulic control signal The deviation from the synchronization angle of the four wheels is as follows: ;

[0016] in To predict the window length, To control the window length, , , and These respectively characterize the steering angle tracking error weight, angular velocity weight, control quantity weight, and synchronization error weight.

[0017] Furthermore, the multi-objective optimization algorithm based on the combined decomposition strategy selects the penalty-based boundary intersection method for solving the multi-objective optimization problem. The normalization function in the algorithm includes:

[0018]

[0019] in This is the ideal point vector for finding the minimum value of each objective. For the weight vector, This is the penalty parameter. For convergent components, For diversity components.

[0020] Furthermore, the convergence component is a characterization Target vector To the ideal point The connection in the weight vector The projection length along the direction is given by the following formula:

[0021]

[0022] The smaller the value, the closer the solution to the optimization problem is to the Pareto front.

[0023] Furthermore, the diversity components are used to characterize Target vector To the ideal point The connection in the weight vector The formula for vertical distance in a direction is as follows:

[0024] The smaller the value, the more the distribution of the solution on the Pareto front conforms to the guidance of the weight vector.

[0025] On the other hand, the specification also provides a data-driven four-wheel multi-objective optimization control device for aerial work platforms, including a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it implements the data-driven four-wheel multi-objective optimization control method for aerial work platforms.

[0026] On the other hand, the specification also provides a computer-readable storage medium on which a program is stored, which, when executed by a processor, implements the aforementioned data-driven four-wheel multi-objective optimization control method for aerial work platforms.

[0027] The beneficial effects of this invention are as follows: This invention proposes a data-driven multi-objective optimization control method for four wheels of an aerial work platform. This method considers the nonlinear characteristics and time-varying influence of hydraulic systems, employs a data-driven pulse neural network for hydraulic system parameter identification, and uses an accurate hydraulic model based on parameter estimation to predict the state at the next moment. Simultaneously, a multi-objective optimization problem is constructed, incorporating the predicted state, reference trajectory, hydraulic control signal, and the deviation of the four-wheel synchronization angle into the optimization problem. The penalty-based boundary intersection method in a decomposition-based multi-objective optimization algorithm is used to solve the multi-objective optimization problem, obtaining the optimal control sequence and achieving synchronous steering control of the four-wheel hydraulic system, minimizing steering deviation. This further reduces problems such as tire wear and travel deviation caused by large angle deviations, thereby improving production efficiency and extending product life. This algorithm takes the four-wheel hydraulic system of an aerial work platform as the research object and introduces multi-objective optimization control to solve the problem of large angle deviations in synchronous steering, which has significant research significance and further exploration value for the high-quality practical application of aerial work platforms. Attached Figure Description

[0028] Figure 1 This is a diagram of a four-wheel multi-objective optimization control framework for an aerial work platform provided in an embodiment of the present invention.

[0029] Figure 2 This is a schematic diagram of the four-wheel structure of the chassis of the aerial work platform provided in an embodiment of the present invention;

[0030] Figure 3 A simplified schematic diagram of hydraulic cylinder-driven tire steering provided in an embodiment of the present invention;

[0031] Figure 4 This is a schematic diagram of the hydraulic principle of a single wheel on the chassis of an aerial work platform provided in an embodiment of the present invention.

[0032] Figure 5A schematic diagram of the four-wheel hydraulic system of an aerial work platform provided in an embodiment of the present invention;

[0033] Figure 6 This is a simulation interface for the four-wheel control of an aerial work platform provided in an embodiment of the present invention.

[0034] Figure 7 The diagram shows the effect of the spiking neural network provided in the embodiment of the present invention;

[0035] Figure 8 This is a diagram illustrating the effect of single-cylinder hydraulic control provided in an embodiment of the present invention.

[0036] Figure 9 A schematic diagram of a data-driven four-wheel multi-objective optimization control device for aerial work platforms provided in an embodiment of the present invention. Detailed Implementation

[0037] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0038] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of this invention.

[0039] like Figure 1 As shown, this invention proposes a data-driven multi-objective optimization control method for four wheels of an aerial work platform. The method primarily focuses on multi-objective optimization control of the four-wheel hydraulic system of the aerial work platform to improve the angle deviation of synchronized steering of the four wheels, thereby increasing operating efficiency and extending equipment lifespan. The specific steps are as follows:

[0040] A four-wheel dynamics model of the chassis of an aerial work platform is constructed. A spiking neural network is used to estimate the time-varying nonlinear parameters in the model, and the four-wheel dynamics model is used as a state prediction model to obtain the state prediction value at the next moment.

[0041] The aerial work platform chassis uses a hydraulic system for steering and drive on all four wheels; therefore, the hydraulic model is constructed as follows:

[0042]

[0043]

[0044]

[0045] in For load traffic, For flow gain, The flow-pressure coefficient, For valve core displacement, For load pressure, For cylinder displacement, The effective area of ​​the hydraulic cylinder piston. It is the total leakage coefficient of the hydraulic cylinder. For the volume of the hydraulic cylinder, This refers to the total mass of the piston and its load, factored onto the piston. For effective bulk modulus, This is the viscous damping coefficient of the piston and load. For the load spring stiffness, This refers to any external load force acting on the piston. When the valve core displacement... With electronic control signals When the relationship between them is linear, ,in This represents the valve-controlled electrical control coefficient.

[0046] like Figure 3 As shown, regarding the displacement of the hydraulic cylinder Rotation angle of hydraulically driven four wheels The relationship is as follows:

[0047]

[0048] Where D is the distance from the bottom of the hydraulic cylinder to the tire flange connection, d is the distance from the top of the hydraulic rod to the point fixed on the tire flange, and H is the length of the hydraulic cylinder. When considering each hydraulic drive system, a subscript needs to be added to each symbol, i.e. , then is used to represent four wheels.

[0049]

[0050]

[0051] in and Let i be the steering angle and steering angular velocity of the i-th wheel. To control the control force of each steering mechanism separately, It is the moment of inertia of the steering mechanism. It is the damping coefficient. It is a nonlinear frictional torque. This is an external disturbance.

[0052] Due to parts of the hydraulic system (e.g., flow gain) Flow pressure coefficient Total leakage coefficient Parameters (such as flow gain) exhibit time-varying nonlinear characteristics, making them difficult to obtain directly. Setting them to fixed values ​​would severely affect the model's dynamic expression and control performance. Therefore, parameters characterizing time-varying nonlinear characteristics (flow gain, etc.) need to be addressed. Flow pressure coefficient Total leakage coefficient This invention utilizes spiking neural networks (SNNs) for dynamic parameter estimation. SNNs simulate the binary pulse information transmission mechanism of the brain, enriching spatiotemporal dynamics and reducing data volume through an event-driven approach. Within the neurodynamic framework of SNNs, the membrane potential equation forms the core mathematical foundation of bio-inspired computation, simulating the essential characteristics of point activity in biological neurons, distinguishing it from the temporal processing capabilities of traditional artificial neural networks. Its membrane potential... Expressed as follows

[0053]

[0054] in The membrane time constant is This is the resting potential. For membrane resistance, This is the synaptic current. When the membrane potential... Exceeding the threshold Then, the neuron will emit a pulse, and the membrane potential will reset to the resting potential. The specific effects are as follows Figure 7 As shown, by using a spiking neural network, the computational load of the neural network is significantly reduced, thereby lowering the computational burden, while the model parameters are dynamically adjusted using a time-triggered mechanism. This allows for the rapid achievement of flow gain using a spiking neural network. Flow pressure coefficient Total leakage coefficient After the time-varying parameters are estimated, the four-wheel dynamics model of the aerial work platform chassis can be constructed into an accurate mathematical model based on parameter estimation. This model can be used for state prediction in model predictive control and to obtain important indicators such as steering angle and steering angular velocity at the next moment.

[0055] Aerial work platform chassis four-wheel synchronous hydraulic control strategy

[0056] This invention employs multi-variable target control for the four wheels of an aerial work platform. By controlling all four wheels with multiple variables, it achieves consistent steering angles across all four wheels and minimizes the angle difference between any two wheels. First, the hydraulic control input sequence is set as follows: ,in To control the window length, an objective function is constructed that only minimizes the predicted state of the four-wheel rotation angle. The steering angle is set with the wheels of the hydraulically driven aerial work platform chassis. The deviation between them, taking into account the hydraulic control signal The deviation between the synchronization angle of the four wheels.

[0057] in To predict the window length, To control the window length, , , and These respectively characterize the steering angle tracking error weight, angular velocity weight, control quantity weight, and synchronization error weight. For a four-wheel hydraulic synchronization system, the following constraints apply:

[0058]

[0059] In the four-wheel control problem of aerial work platforms, there are control variables for four wheels. This study introduces a decomposition-based multi-objective optimization algorithm to solve the optimal control sequence for the four wheels of the aerial work platform. This algorithm addresses the multi-objective optimization problem for the four wheels of the aerial work platform using a set of uniformly distributed weight vectors. Decompose into several subproblems For each subproblem objective function, the form is as follows:

[0060]

[0061] Then, an evolutionary algorithm is used to find the optimal solution for the subproblems of each wheel. Cooperative methods such as local neighbor cooperation, information sharing, and reference point cooperation are used to optimize the subproblems of each wheel, thereby achieving optimal four-wheel control of the aerial work platform globally. This paper introduces a penalty strategy in the boundary intersection method. The specific steps are as follows: First, a standardized function is constructed.

[0062] Min

[0063]

[0064] in This is the ideal point vector for finding the minimum value of each objective. For the weight vector, The penalty parameter is used. First, a weight vector is randomly generated. The neighbor structure calculates the Euclidean distance between the weight vector and other vectors, selects several nearest neighbors, and calculates the objective function. The initial ideal point vector is obtained. Each dimension of the reference point represents the minimum value of the corresponding objective function among all solutions. Then, neighbors are randomly selected and crossover mutations are performed to generate new individuals. An evolutionary algorithm is then used to solve the objective function of the new individuals, updating the ideal point vector. This involves updating neighbor solutions and the Pareto front, thereby achieving the final optimal solution. For convergent components, characterizing Target vector To the ideal point The connection in the weight vector The projected length along the direction. See below for details:

[0065]

[0066] The smaller the value, the closer the solution to the optimization problem is to the Pareto front. For the diversity components, express Target vector To the ideal point The connection in the weight vector Vertical distance in the direction. See below for details:

[0067]

[0068] Characterizes the degree of inverse deviation between the solution and the specified weights. The smaller the value, the more the distribution of solutions on the Pareto front conforms to the guidance of the weight vector. The optimal solution is achieved by solving for the weight vector, where each solution is as close as possible to its corresponding reference line and evenly distributed along this reference line.

[0069] After transforming the multi-objective optimization problem into multiple single-objective problems and solving them, an optimal control sequence for the four wheels of the aerial work platform is formulated. This ensures effective control of the hydraulic system of the four wheels at the current moment, guarantees that the deviation range of the synchronous steering process of the four wheels is controllable, and after multiple iterations and tests of the system, the synchronous steering deviation of the controlled four-wheel system is finally minimized. Figure 8 Given the set conditions, the hydraulic system can track and eventually eliminate the error to zero, effectively avoiding the problem of excessive final error in the four-wheel system.

[0070] Method simulation verification and deployment

[0071] After theoretical testing and verification of the method, this invention aims to construct a joint simulation model of the four-wheel hydraulic control of the aerial work platform chassis using Simulink and AMEsim. The four-wheel chassis structure of the aerial work platform is as follows: Figure 2 As shown, it specifically includes an electric motor and diesel fuel pump for the oil supply section, as well as a four-way hydraulic system for the tires, wherein the hydraulic system is as follows: Figure 4 and Figure 5 As shown, it includes hydraulic valves, hydraulic cylinders, and pipelines. (Combined with...) Figure 5 The aerial work platform's four-wheel chassis and four-wheel hydraulic system are jointly simulated with the four-wheel chassis mechanical structure diagram in Simulink (e.g., Figure 6 In Simulink, a hydraulic control algorithm for four wheels is constructed and deployed in the real-time control system of the aerial work platform for synchronous control of the four wheels.

[0072] Corresponding to the aforementioned embodiment of a data-driven four-wheel multi-objective optimization control method for aerial work platforms, the present invention also provides an embodiment of a data-driven four-wheel multi-objective optimization control device for aerial work platforms.

[0073] See Figure 9 The present invention provides a data-driven four-wheel multi-objective optimization control device for aerial work platforms, comprising a memory and one or more processors. The memory stores executable code, and when the processor executes the executable code, it is used to implement a data-driven four-wheel multi-objective optimization control method for aerial work platforms as described in the above embodiment.

[0074] The present invention provides an embodiment of a data-driven four-wheel multi-objective optimization control device for aerial work platforms. This device can be applied to any device with data processing capabilities, such as a computer. The device embodiment can be implemented through software, hardware, or a combination of both. Taking software implementation as an example, as a logical device, it is formed by the processor of any data-processing device loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, such as... Figure 9 The diagram shown is a hardware structure diagram of any device with data processing capabilities, including a data-driven four-wheel multi-objective optimization control device for aerial work platforms provided by this invention. (Except for...) Figure 9 In addition to the processor, memory, network interface, and non-volatile memory shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0075] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0076] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0077] This invention also provides a computer-readable storage medium storing a program thereon, which, when executed by a processor, implements a data-driven four-wheel multi-objective optimization control method for aerial work platforms as described in the above embodiments.

[0078] The computer-readable storage medium can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device of any data processing device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices of any data processing device. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0079] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the aforementioned data-driven four-wheel multi-objective optimization control method for aerial work platforms.

[0080] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0081] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this application. This application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A data-driven aerial work platform four-wheel multi-objective optimization control method, characterized in that, The method comprises: A four-wheel dynamics model of the aerial work platform chassis is constructed, time-varying nonlinear parameters in the model are estimated using a pulse neural network, and a state prediction value at the next time is obtained using the four-wheel dynamics model as a state prediction model; A hydraulic control sequence is constructed, a target function is constructed with the aim of minimizing the deviation between the prediction value and the set steering angle of the wheels of the aerial work vehicle chassis, a multi-objective optimization algorithm based on a combination decomposition strategy is used to decompose the four-wheel optimization problem into single-wheel optimization problems, multiple single-wheel optimization problems are simultaneously optimized, and the optimal hydraulic control sequence of the four wheels is obtained.

2. The data-driven aerial work platform four-wheel multi-objective optimization control method according to claim 1, characterized in that, The four-wheel dynamics model of the aerial work platform chassis is specifically a hydraulic system: wherein is the load flow, is the flow gain, is the flow pressure coefficient, is the spool displacement, is the load pressure, is the cylinder displacement, is the hydraulic cylinder piston effective area, is the hydraulic cylinder total leakage coefficient, is the hydraulic cylinder volume, is the total mass of the piston and load reduced to the piston, is the effective volume elastic modulus, is the viscous damping coefficient of the piston and load, K is the load spring stiffness, is the arbitrary external load force acting on the piston.

3. The data-driven aerial work platform four-wheel multi-objective optimization control method of claim 1, wherein, The time-varying nonlinear parameters include flow gain , flow pressure coefficient , and total leakage coefficient of the hydraulic cylinder .

4. The data-driven aerial work platform four-wheel multi-objective optimization control method of claim 1, wherein, The membrane potential equation of the pulse neural network is expressed as follows ； wherein is the membrane time constant, is the resting potential, is the membrane resistance, is the synaptic current, when the membrane potential V exceeds the threshold After, the neuron fires a spike, the membrane potential is reset to the resting potential ; the model parameters of the spiking neural network are dynamically adjusted with a time trigger mechanism.

5. The data-driven aerial work platform four-wheel multi-objective optimization control method of claim 1, wherein, The objective function minimizes the deviation between the predicted states of the four wheel turning angles The deviation between the hydraulic control signals and the set steering angles of the wheels of the aerial vehicle chassis is taken into account The deviation between the hydraulic control signals and the set steering angles of the wheels of the aerial vehicle chassis is taken into account The deviation between the hydraulic control signals and the set steering angles of the wheels of the aerial vehicle chassis is taken into account ; wherein is the prediction window length, is the control window length, and are the steering angle tracking error weight, the angular velocity weight, the control weight and the synchronization error weight, respectively.

6. The data-driven aerial work platform four-wheel multi-objective optimization control method of claim 1, wherein, The multi-objective optimization algorithm based on the combination decomposition strategy selects a penalty-based boundary intersection method for multi-objective optimization solution, and the standardization function in the algorithm comprises: ； wherein is the ideal point vector currently found for each target minimum, is the weight vector, is the penalty parameter, is the convergence component, is the diversity component.

7. The data-driven aerial work platform four-wheel multi-objective optimization control method according to claim 6, characterized in that, The convergence component is the projection length of the line connecting the x target vector to the ideal point in the direction of the weight vector , which is given by the formula: ； The smaller the value, the closer the solution to the optimization problem is to the Pareto front.

8. The data-driven aerial work platform four-wheel multi-objective optimization control method according to claim 7, characterized in that, The diversity component is characterized by Target vector The perpendicular distance from the ideal point to the weight vector in the direction of the target vector, is given by the formula: ; The smaller, the more the distribution of solutions on the Pareto front conforms to the guidance of the weight vector.

9. A data-driven four-wheel multi-objective optimization control device for aerial work platforms, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that... The processor executes the executable code to implement the data-driven four-wheel multi-objective optimization control method of the aerial work platform as claimed in any one of claims 1-8.

10. A computer-readable storage medium having stored thereon a program, characterized in that, The program is executed by the processor to implement the data-driven four-wheel multi-objective optimization control method of the aerial work platform as claimed in any one of claims 1-8.

Images