Electric turntable adaptive variable speed control system and method based on BP neural network
By applying BP neural network on the electric turntable for adaptive speed control, the safety and efficiency of equipment at fixed speeds are solved, and smooth operation and efficient control are achieved.
Patent Information
- Application Number
- CN202411776583.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-05
- Publication Date
- 2025-04-11
AI Technical Summary
The existing electric rotary table uses a fixed speed during startup-operation-stop, which cannot take into account both the safety and working efficiency of the equipment. Especially in the case of heavy load and large volume, there may be problems such as shaking, tilting or even losing balance.
The electric rotary table adaptive speed control system based on BP neural network is adopted to monitor the load in real time through pressure sensors, and the BP neural network model is used to dynamically adjust the speed according to parameters such as load, angle, acceleration time and deceleration time.
It realizes high-precision control of the electric rotary table, ensures smooth start, operation and stop, improves the service life and safety of the equipment, reduces the risk of equipment damage, and improves work efficiency.
Smart Images
Figure CN119806225A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of automatic control equipment, and relates to an electric turntable adaptive speed control system based on a BP neural network. The invention also relates to an electric turntable adaptive speed control method based on a BP neural network. Background Art
[0002] Existing electric turntables use a fixed speed during the start-run-stop process. When testing products such as loudspeakers in an outdoor environment, the relevant test standards require that the center of the sound source should be more than 1.5 meters above the ground. When the load is heavy and the volume is large, the turntable may shake, tilt, or even lose balance and overturn during the start-run-stop process, causing damage to the turntable and the product, and may also pose a safety threat to the user. At the same time, if the speed is set too slow during the operation stage, it will reduce work efficiency.
[0003] Therefore, there is an urgent need to develop a solution that can dynamically adjust the rotation speed according to the real-time load conditions, which can both ensure test safety and improve work efficiency. Summary of the invention
[0004] The purpose of the present invention is to provide an electric turntable adaptive speed control system based on BP neural network, which solves the problem that the existing electric turntable adopts a fixed speed during the start-run-stop process and cannot take into account both equipment safety and work efficiency.
[0005] Another object of the present invention is to provide an adaptive speed control method for an electric turntable based on a BP neural network, which solves the problem that the turntable may shake, tilt or even lose balance during the start-run-stop process when the load is heavy and the volume is large.
[0006] The technical solution adopted by the present invention is an adaptive speed control system for an electric turntable based on a BP neural network, comprising a bracket supporting a lower bottom plate, the lower bottom plate supporting an upper bottom plate upward, a hollow turntable arranged between the upper bottom plate and the lower bottom plate, a turntable arranged on the upper surface of the upper bottom plate, a level meter arranged on the turntable, the turntable is drivingly connected to the hollow turntable, and the hollow turntable is equipped with a servo motor; The servo motor is connected to the central processing unit through a servo driver; a fixed plate is supported on the turntable through a plurality of pressure sensors, signal lines of the pressure sensors are connected to the central processing unit, and the servo driver and the central processing unit are both connected to the control panel.
[0007] Another technical solution adopted by the present invention is an electric turntable adaptive speed control method based on BP neural network, which is implemented by using the above-mentioned electric turntable adaptive speed control system based on BP neural network according to the following steps: Step 1: Data collection: Before operation, the load weight of the turntable on the electric turntable is monitored in real time through the pressure sensor; at the same time, the command information of the rotation angle of the electric turntable is input through the control panel; Step 2: Network construction, training and calculation: 2.1) Use historical data sets to build BP neural network models. 2.2) Complete the training and calculation of the BP neural network model; Step 3: Run in real time based on the calculated output and error.
[0008] The beneficial effect of the present invention is that the BP neural network is used to design the adaptive speed control of the electric turntable. This method can input multiple real-time parameters (such as load, angle, acceleration time, running time, deceleration time, etc.) to autonomously adjust the control strategy, which not only realizes the high-precision control of the electric turntable, but also realizes the smooth start, operation and stop of the electric turntable. The use effect is good, the service life of the electric turntable is increased, the safety of personnel and equipment is guaranteed, the cost is low, and the online upgrade and maintenance are simple. The control system of the present invention can respond quickly when facing sudden load changes, ensure smooth start, operation and stop under different load conditions, improve work efficiency and reduce the risk of equipment damage. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 It is a schematic diagram of the structure adopted by the control system of the present invention; Figure 2 It is the training and calculation process of the BP neural network model adopted by the control system of the present invention; Figure 3 It is a schematic diagram of the BP neural network model adopted by the control system of the present invention.
[0010] In the figure, 1. level, 2. turntable, 3. servo motor, 4. servo drive controller, 5. anchor bolt, 6. bracket, 7. fixable pulley, 8. fixing plate, 9. pressure sensor, 10. benchmark, 11. upper base plate, 12. hollow rotating table, 13. support column, 14. central processing unit, 15. control panel, 16. lower base plate. DETAILED DESCRIPTION
[0011] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.
[0012] In existing research, acceleration and deceleration time is generally not used as an evaluation indicator due to problems such as prediction accuracy and speed. Taking acceleration and deceleration time into consideration significantly increases decision-making information, which helps to make a more accurate target allocation plan. At the same time, in complex practical application environments, the acceleration and deceleration time of the electric turntable is an important, real-time factor. Incorporating it into the target allocation considerations helps to dynamically adjust the allocation strategy and optimize practical effects. The control system of the present invention is based on a neural network. By adding an electric turntable, the data set generated by the simulation experiment of the motion posture under different load conditions is trained to obtain a prediction model for the start-run-stop speed that can take into account both real-time and accuracy. The prediction results are closer to real-world application scenarios than traditional control methods.
[0013] like Figure 1 As shown, the hardware structure relied on by the control system of the present invention includes a bracket 6 equipped with a fixed pulley 7 (with a self-locking function), a lower base plate 16 is supported on the upper surface of the bracket 6 by a plurality of anchor bolts 5, the lower base plate 16 is fixedly supported by an upper base plate 11 upwardly by four support columns 13, a hollow rotating table 12 is arranged in the space between the upper base plate 11 and the lower base plate 16, a turntable 2 is arranged at the center position of the upper surface of the upper base plate 11, a level 1 is also arranged on the turntable 2, the level 1 corresponds to the benchmark 10 on the upper base plate 11, and the turntable 2 is connected to the hollow rotating table 12 by driving; the hollow rotating table 12 is equipped with a servo motor 3, and the servo motor 3 is connected to the central processing unit 14 through a servo drive controller 4; a fixed plate 8 is supported and fixed on the turntable 2 by a plurality of pressure sensors 9, the signal line of the pressure sensor 9 is connected to the central processing unit 14, and the servo motor 3 and the central processing unit 14 are both connected to the control panel 15 (which may include a touch screen) by signal.
[0014] The functions of each component are described as follows: 1) Pressure sensor 9: measures the load weight on the fixed plate 8 of the electric turntable in real time. The number of pressure sensors 9 is set to 2 to 4; 2) Central Processing Unit 14 (CPU): a BP neural network model is pre-stored therein, which receives the load data collected by the pressure sensor 9, and takes the arithmetic average of the load data collected by multiple pressure sensors 9; executes the adaptive speed change algorithm in the BP neural network model, predicts the speed of the electric turntable through the adaptive speed change algorithm table, and outputs a command signal of the turntable speed according to the prediction result; 3) Servo drive controller 4: performs specific state control on servo motor 3 according to the command signal of turntable speed output by central processor 14, so as to realize real-time adjustment of the running state of electric turntable; 4) Control panel 15 and touch screen: The operator manually inputs the rotation angle and other parameters of the target position, including start and stop control.
[0015] 5) A voice recognition module may also be added and connected to the central processor 14 to achieve functional expansion and more convenient and detailed control.
[0016] Reference Figure 3 , the central processing unit pre-stores a BP neural network model, and the architecture of the BP neural network model includes the following three layers: a) Input layer: contains three input nodes, corresponding to the real-time load weight , target rotation angle and acceleration ( ); b) Hidden layer: There are multiple hidden layers, and the number of neurons is adjusted according to actual needs to extract data features; c) Output layer: There are two output nodes, which output the starting speed signal of the adaptive variable speed electric turntable control. , stop speed signal .
[0017] The working principle of the control system of the present invention is: A plurality of pressure sensors are installed between the fixed plate and the turntable of the existing electric turntable, and the output ends of the plurality of pressure sensors are connected to the signal interface of the central processing unit through signal lines. The plurality of pressure sensors upload the load weight information on the turntable to the central processing unit; the central processing unit models and creates an adaptive speed change algorithm based on the inertia principle, and uses the adaptive speed change algorithm to establish a rotation control mode according to the load weight and the rotation angle command input by the control panel as well as the acceleration time and the deceleration time. The values of the starting speed and the ending speed are calculated through the rotation control mode, and the central processing unit then controls the corresponding work of the servo motor through the servo drive controller until the turntable of the electric turntable finally stops smoothly at the specified position.
[0018] Table 1. Adaptive speed change algorithm
[0019] As shown in Table 1 and Table 2, it is a preset adaptive speed change algorithm table, which is used to query and preset the expected starting speed, running speed and ending speed according to the load weight and rotation angle instruction. The design description of the adaptive speed change algorithm table is as follows: 1) Load weight : The turntable load weight monitored in real time by the pressure sensor, (kg); 2) Rotation angle : The desired rotation angle input by the user through the control panel, (°); 3) Starting speed : Initial speed setting under specific load and angle command, (r / min); 4) Operating speed : After starting, the speed is set to maintain stable operation; 5) End speed : The target speed for smooth deceleration to ensure safe stopping of the equipment; 6) Acceleration time : The time required from the motor starting speed to the running speed, (s); 7) Running time : Time of uniform rotation, (s).
[0020] 8) Deceleration time : The time required from the running speed to the end speed, (s).
[0021] The hardware structure of the control system of the present invention supports online upgrading, and the user can retrain the BP neural network model in the central controller according to actual usage to adapt to new working conditions or load changes.
[0022] Table 2: Adaptive speed change algorithm
[0023] The core of the BP neural network model is to use an adaptive speed-changing algorithm, which is described as follows: According to the data collected by the pressure sensor in real time, the BP neural network model optimizes the weights and thresholds through the BP learning process of forward propagation and back propagation, thereby calculating the turntable speed under a given load and target angle; The BP learning process consists of forward propagation and backward propagation, in which the input signal of the forward propagation is transmitted from the input layer to the output layer through the hidden layer; the speed value obtained by the output layer is assigned to the servo drive controller, and the CPU calculates the error of multiple pressure sensor signals collected in real time according to step 2.2.5). If the error is less than the threshold, it is confirmed that the speed output by the output layer at this moment is the expected speed, and the BP learning process ends; otherwise, it is converted to backward propagation. The forward propagation is to reversely calculate the error signal according to the original connection path, and adjust the weights and thresholds of each layer of neurons by the gradient descent method to reduce the error signal to an allowable range.
[0024] Note: 1. Input angle is the turntable rotation angle; 2. Start, run and end speeds are all motor speeds. For example, when the reduction ratio is 100, the motor speed of 100r / min is equivalent to the turntable speed of 6° / s.
[0025] The electric turntable adaptive speed control method based on BP neural network of the present invention is implemented according to the following steps: Step 1: Data collection: Before operation, the load weight of the turntable on the electric turntable is monitored in real time through the pressure sensor; at the same time, the command information of the rotation angle of the electric turntable is input through the control panel (touch screen); Step 2: Network construction, training and calculation: 2.1) Use historical data sets to build BP neural network models. The data in the historical data set include the best speed records under different loads and angles; The training process includes: Forward propagation: calculate the output of each layer until the output layer; Error calculation: calculate the difference between the output result and the actual optimal speed; Back propagation: adjust the network weights and thresholds according to the error, complete the training, and obtain the optimized BP neural network model; Input layer: contains three input nodes, namely real-time load weight and acceleration ; Hidden layer: There are multiple hidden layers, including three neurons; Output layer: There are two output nodes, which output two speed signals, namely the motor starting speed and ending speed.
[0026] 2.2) Complete the training and calculation of the BP neural network model, refer to Figure 2 , the process is as follows: 2.2.1) Set variables and parameters, perform normalization, and set the input vector (training sample) to X l ( k )=[ x l1 ( k ), x l2 ( k ), x l3 ( k )],in l =1,2,…,K, K is the number of training samples; x l1 ( k ), x l2 ( k ), x l3 ( k )] represent the input parameters during the start, run and stop process respectively; represents the acceleration of the electric turntable, Indicates the operating angle of the electric turntable. Indicates the current load of the electric turntable; According to the principle of automatic speed change, select The three parameters are used as input variables of the BP neural network model, and the output is two speed signals. Figure 3 As shown; 2.2.2) Settings is the weight vector between the input layer and the hidden layer at the kth iteration; set is the weight vector between the hidden layer and the output layer at the kth iteration; set up Y l ( k ), = [ y l1 ( k ), y l2 ( k )] are two speed signals of the network at the kth iteration, l=1,2,…,K; corresponding to the adaptive speed control, y l1 ( k ), y l2 ( k ) respectively represent two speed signals of the adaptive speed control system; 2.2.3) Initialization: Assign , Each of them has a small random non-zero value to provide an initial weight for learning; at the same time, set the safety monitoring value: the maximum load is 90kg; 2.2.4) Forward calculation: For random input samples X l , forward calculation of the input signal and output signal of each layer of neurons in the BP network, the expression is: (1) (2) (3) (4) (5) in, For the hidden layer i The input signal of each neuron; N is the number of neurons in the input layer; The input layer n neurons and hidden layers i The connection weights of neurons; Indicates the input parameters of the electric turntable during start, operation and stop; l is the number of hidden layer neurons; I Enter the number of samples; For the hidden layer i The output signal of a neuron; f(U) is the Sigmoid function; The output layer j The input signal of each neuron; The input layer n neurons and hidden layers i The connection weights of neurons; J is the number of neurons in the output layer; is the output signal of the jth neuron in the output layer; k Indicates the number of iterations; ( k ) indicates the k The actual output of the iteration.
[0027] 2.2.5) If the data collected by the four pressure sensors 9 are: , by comparing the maximum and minimum values are: , = ; The control system automatically adjusts the speed when the load exceeds the threshold to avoid loss of balance. During the speed adjustment process, the control system will continuously track the status of the electric turntable to ensure stable operation and prevent abnormal situations.
[0028] The threshold is obtained by calculation , , according to the allowable error range of engineering mechanics, determine whether the error meets the preset allowable range (for example, whether the safety threshold is less than 0.05); if it does, go to step 2.2.8); if it does not, go to step 2.2.6); 2.2.6) Determine whether k+1 is greater than the maximum number of iterations 100. If so, go to step 2.2.7); if not, perform the input sample X l ( k ) reversely calculate the local gradient δ of each layer of neurons, expressed as: (6) (7) 2.2.7) Calculate weight correction Δw , and correct the weights; k = k +1, go to step 2.2.4), the expression is: (8) (9) (10) (11) In the formula, is the learning rate; 2.2.8) Determine whether all training samples have been learned. If yes, end and go to step 3; otherwise, go to step 2.2.3) until E(k) until the threshold range is met.
[0029] Step 3: Run in real time based on the calculated output and error: 3.1) During operation, each pressure sensor obtains load and target angle data in real time; 3.2) The real-time collected data of each pressure sensor and target angle are transmitted to the central controller, which uses the optimized BP neural network model to calculate the error values of the three stages of smooth start, operation and stop under the current conditions. If the threshold range is met, the optimal speed values are sent to the servo drive controller; 3.3 The servo drive controller dynamically adjusts the speed of the electric turntable according to the received output results to ensure smooth transition during starting, running and stopping.
[0030] Example 1 The control system of the present invention is constructed according to the aforementioned hardware structure. The electric turntable adaptive speed control system based on the BP neural network includes a bracket 6 with a lower bottom plate 16 supported on the upper surface, the lower bottom plate 16 supports an upper bottom plate 11 upward, a hollow turntable 12 is arranged in the space between the upper bottom plate 11 and the lower bottom plate 16, a turntable 2 is arranged at the center position of the upper surface of the upper bottom plate 11, a level 1 is arranged on the turntable 2, the turntable 2 is drivingly connected to the hollow turntable 12, and the hollow turntable 12 is equipped with a servo motor 3; The servo motor 3 is connected to the central processing unit 14 through the servo drive controller 4; a fixed plate 8 is supported and fixed on the turntable 2 through multiple pressure sensors 9, and the signal lines of the pressure sensors 9 are connected to the central processing unit 14. The servo motor 3 and the central processing unit 14 are both connected to the control panel 15 signal.
[0031] Pressure sensor: Two are set to measure the load weight on the turntable in real time.
[0032] The central processor 14, which pre-stores a BP neural network model, receives the load data collected by the pressure sensor 9, and takes the average value of the load data collected by multiple pressure sensors 9; executes the adaptive speed change algorithm in the BP neural network model, and predicts and calculates the optimal speed of the electric turntable through the adaptive speed change algorithm table, and outputs the optimal speed command signal according to the prediction result.
[0033] Example 2 The control system of the present invention is constructed according to the aforementioned hardware structure. The electric turntable adaptive speed control system based on the BP neural network includes a bracket 6 with a lower bottom plate 16 supported on the upper surface, the lower bottom plate 16 supports an upper bottom plate 11 upward, a hollow turntable 12 is arranged in the space between the upper bottom plate 11 and the lower bottom plate 16, a turntable 2 is arranged at the center position of the upper surface of the upper bottom plate 11, a level 1 is arranged on the turntable 2, the turntable 2 is drivingly connected to the hollow turntable 12, and the hollow turntable 12 is equipped with a servo motor 3; The servo motor 3 is connected to the central processing unit 14 through the servo drive controller 4; a fixed plate 8 is supported and fixed on the turntable 2 through multiple pressure sensors 9, and the signal lines of the pressure sensors 9 are connected to the central processing unit 14. The servo motor 3 and the central processing unit 14 are both connected to the control panel 15 signal.
[0034] Pressure sensors, three in number, are installed at different positions of the turntable to measure the load weight more accurately.
[0035] The BP neural network model has the following three-layer structure: a) Input layer: contains three input nodes, corresponding to the real-time load weight and target rotation angle , acceleration ; b) Hidden layer: There are multiple hidden layers to extract data features; c) Output layer: There are two output nodes, which output the starting speed signal of the adaptive variable speed electric turntable control. y l1 ( k ), stop speed signal y l2 ( k ).
[0036] Example 3 The control system of the present invention is constructed according to the aforementioned hardware structure. The electric turntable adaptive speed control system based on the BP neural network includes a bracket 6 with a lower bottom plate 16 supported on the upper surface, the lower bottom plate 16 supports an upper bottom plate 11 upward, a hollow turntable 12 is arranged in the space between the upper bottom plate 11 and the lower bottom plate 16, a turntable 2 is arranged at the center position of the upper surface of the upper bottom plate 11, a level 1 is arranged on the turntable 2, the turntable 2 is drivingly connected to the hollow turntable 12, and the hollow turntable 12 is equipped with a servo motor 3; The servo motor 3 is connected to the central processing unit 14 through the servo drive controller 4; a fixed plate 8 is supported and fixed on the turntable 2 through multiple pressure sensors 9, and the signal lines of the pressure sensors 9 are connected to the central processing unit 14. The servo motor 3 and the central processing unit 14 are both connected to the control panel 15 signal.
[0037] On the basis of Example 1, a photoelectric sensor is additionally provided to monitor the rotation speed of the electric turntable.
[0038] Photoelectric sensor: monitors the rotation speed of the electric turntable in real time and transmits data to the central processor to affect the adaptive speed control strategy.
[0039] BP neural network model: A node is added to the input layer, which contains real-time speed information to optimize the speed output.
[0040] Example 4 The control system of the present invention is constructed according to the aforementioned hardware structure. The electric turntable adaptive speed control system based on the BP neural network includes a bracket 6 with a lower bottom plate 16 supported on the upper surface, the lower bottom plate 16 supports an upper bottom plate 11 upward, a hollow turntable 12 is arranged in the space between the upper bottom plate 11 and the lower bottom plate 16, a turntable 2 is arranged at the center position of the upper surface of the upper bottom plate 11, a level 1 is arranged on the turntable 2, the turntable 2 is drivingly connected to the hollow turntable 12, and the hollow turntable 12 is equipped with a servo motor 3; The servo motor 3 is connected to the central processing unit 14 through the servo drive controller 4; a fixed plate 8 is supported and fixed on the turntable 2 through multiple pressure sensors 9, and the signal lines of the pressure sensors 9 are connected to the central processing unit 14. The servo motor 3 and the central processing unit 14 are both connected to the control panel 15 signal.
[0041] Based on Example 1, the control panel function is enhanced by adding a voice recognition module to allow operators to input target positions and parameters by voice.
[0042] Voice recognition module: recognizes the operator's voice commands and converts them into control signals to facilitate barrier-free operation.
[0043] Central Processing Unit: Processes input from voice recognition and touch screen to achieve more flexible control methods.
[0044] Example 5 The adaptive speed control method based on the BP neural network algorithm of the present invention is implemented according to the following steps: Step 1: Data collection, real-time load weight =30kg, target rotation angle is =90 degrees, calculate the starting acceleration , stop acceleration :
[0045]
[0046] Step 2: Network construction, training and calculation: Setting of BP neural network model: Input layer: contains three input nodes, acceleration , real-time load weight , target rotation angle ; Hidden layer: There are two hidden layers, containing three neurons; Output layer: There are two output nodes, which output the motor start speed and stop speed signals; Weight initialization: Assume that the weight between the input layer and the hidden layer is: W1=[0.1,0.2,0.3], calculate the input signal: = , .
[0047] Apply activation function (assuming sigmoid): , .
[0048] Step 3: Based on the calculated output and error, Expected output: The threshold of the expected output E(k)<5% , Actual output: Assuming that the pressure sensor is used L =[7.45,7.51,7.43,7.61], calculate the threshold, , Calculation error: 2.4%<5%, indicating that the output speed meets the experimental requirements, that is, the turntable starting speed is 33.058r / min, and the turntable stopping speed is 32.943r / min.
[0049] Example 6 The adaptive speed control method based on the BP neural network algorithm of the present invention is implemented according to the following steps: Step 1: Data collection, real-time load weight =45kg, target rotation angle is =150 degrees, calculate the starting acceleration , stop acceleration : , .
[0050] Step 2: Network training and calculation, Setting of BP neural network model: Input layer: The three input nodes are, acceleration , real-time load weight , target rotation angle ; Hidden layer: There are two hidden layers, containing three neurons; Output layer: Two output nodes are set to output the turntable motor starting speed and motor ending speed signals.
[0051] Weight initialization: Assume that the weight between the input layer and the hidden layer is: W1=[0.15,0.25,0.35], Calculate the input signal:
[0052] .
[0053] Apply activation function (assuming sigmoid): , .
[0054] Step 3: Calculate the expected output and threshold error based on . Expected output: The threshold of the expected output E(k)<5% ; Actual output: Assuming that the pressure sensor is used L =[11.68,11.05,11.27,11], calculate the threshold, , Calculation error: 6.04% > 5%, indicating that the first output speed does not meet the experimental requirements. Since the number of iterations k+1<100, the weights and thresholds of neurons in each layer are adjusted according to the gradient descent method in step 2.2.7), and recalculation is required.
Claims
1. The electric turntable adaptive speed control system based on BP neural network is characterized by: The device comprises a bracket (6) supporting a lower base plate (16), the lower base plate (16) supporting an upper base plate (11) upwards, a hollow rotating platform (12) being arranged between the upper base plate (11) and the lower base plate (16), a rotating disk (2) being arranged on the upper surface of the upper base plate (11), a level (1) being arranged on the rotating disk (2), the rotating disk (2) being drivingly connected to the hollow rotating platform (12), and the hollow rotating platform (12) being equipped with a servo motor (3); The servo motor (3) is connected to the central processing unit (14) via a servo drive controller (4); a fixed plate (8) is supported and fixed on the turntable (2) via a plurality of pressure sensors (9); signal lines of the pressure sensors (9) are connected to the central processing unit (14); and both the servo motor (3) and the central processing unit (14) are connected to the control panel (15) via signals.
2. The electric turntable adaptive speed control system based on BP neural network according to claim 1 is characterized in that: The central processor (14) has a BP neural network model pre-stored therein, receives load data collected by the pressure sensor (9), takes an average value of the load data collected by multiple pressure sensors (9), executes the adaptive speed change algorithm in the BP neural network model, performs prediction calculation of the optimal speed of the electric turntable through the adaptive speed change algorithm table, and outputs an optimal speed command signal according to the prediction result.
3. The electric turntable adaptive speed control system based on BP neural network according to claim 2 is characterized in that: The BP neural network model has the following three-layer structure: a) Input layer: contains three input nodes, corresponding to the acceleration of the servo motor (3) , the real-time load weight L of the fixing plate (8), the rotation angle of the fixing plate (8) ; b) Hidden layer: There are multiple hidden layers, and the number of neurons is adjusted according to actual needs to extract data features; c) Output layer: Two output nodes are set to output two speed signals, namely the motor starting speed and ending speed.
4. The electric turntable adaptive speed control system based on BP neural network according to claim 2 is characterized in that: The adaptive speed change algorithm table is used to convert the load weight and rotation angle command into the optimal motor starting speed, running speed and motor ending speed, as described below: 1) Load weight : The turntable load weight is monitored in real time by the pressure sensor; 2) Rotation angle : The desired rotation angle input by the user through the control panel; 3) Starting speed : Initial speed setting under specific load and angle command; 4) Operating speed : After starting, the speed is set to maintain stable operation; 5) End speed : The target speed for smooth deceleration to ensure safe stopping of the equipment; 6) Acceleration time : The time required from the motor starting speed to the running speed; 7) Running time : time of uniform rotation; 8) Deceleration time : The time required from the running speed to the end speed.
5. The electric turntable adaptive speed control system based on BP neural network according to claim 1 is characterized in that: The servo drive controller (4) performs specific state control on the servo motor (3) according to the turntable speed command signal output by the central processing unit (14), thereby achieving real-time adjustment of the operating state of the electric turntable.
6. An adaptive speed control method for an electric turntable based on a BP neural network, using the adaptive speed control system for an electric turntable based on a BP neural network as claimed in any one of claims 1 to 5, characterized in that: Follow these steps to implement: Step 1: Data collection: Before operation, the load weight of the turntable on the electric turntable is monitored in real time through the pressure sensor; at the same time, the command information of the rotation angle of the electric turntable is input through the control panel; Step 2: Network construction, training and calculation: 2.1) Use historical data sets to build BP neural network models. 2.2) Complete the training and calculation of the BP neural network model; Step 3: Run in real time based on the calculated output and error.
7. The method for adaptive speed control of an electric turntable based on a BP neural network according to claim 6 is characterized in that: In step 2.1), the specific process is: The historical data set contains the best speed records under different loads and angles. The training process includes: Forward propagation: calculate the output of each layer until the output layer; Error calculation: calculate the difference between the output result and the actual optimal speed; Back propagation: adjust the network weights and thresholds according to the error, complete the training, and obtain the optimized BP neural network model; Setting of BP neural network model: Input layer: contains multiple input nodes, including acceleration , real-time load weight , target rotation angle ; Hidden layer: There are multiple hidden layers, including three neurons; Output layer: There are two output nodes, which output two speed signals.
8. The method for adaptive speed control of an electric turntable based on a BP neural network according to claim 6 is characterized in that: In step 2.2), the specific process is: 2.2.1) Set variables and parameters, perform normalization, and set the input vector to X l ( k )=[ x l1 ( k ), x l2 ( k ), x l3 ( k )],in l =1,2,…,K, K is the number of training samples; x l1 ( k ), x l2 ( k ), x l3 ( k )] represent the input parameters during the start, run and stop process respectively; 2.2.2) Settings is the weight vector between the input layer and the hidden layer at the kth iteration; set is the weight vector between the hidden layer and the output layer at the kth iteration; 2.2.3) Initialization: Assign , Each of them has a small random non-zero value; at the same time, a safety monitoring value is set; 2.2.4) Forward calculation: For random input samples X k , forward calculation of the input signal of each layer of neurons in the BP network U and output signal V ; 2.2.5) Calculate the threshold value from the data collected by the pressure sensor, and judge whether the error meets the preset allowable range according to the allowable error range of engineering mechanics. If so, go to step 2.2.8); if not, go to step 2.2.6); 2.2.6) Determine whether k+1 is greater than the maximum number of iterations 100. If so, go to step 2.2.7); if not, perform the input sample X l ( k ) Reversely calculate the local gradient δ of each layer of neurons; 2.2.7) Calculate weight correction Δw , and correct the weights; k = k +1, go to step 2.2.4); 2.2.8) Determine whether all training samples have been learned. If yes, go to step 3; otherwise, go to step 2.2.3) until E(k) until the threshold range is met.
9. The method for adaptive speed control of an electric turntable based on a BP neural network according to claim 6, characterized in that: In step 3, the specific process is: 3.1) During operation, each pressure sensor obtains load and target angle data in real time; 3.2) The real-time collected data of each pressure sensor and target angle are transmitted to the central controller, which uses the optimized BP neural network model to calculate the error values of the three stages of smooth start, operation and stop under the current conditions. If the threshold range is met, the optimal speed values are sent to the servo drive controller; 3.3 The servo drive controller dynamically adjusts the speed of the electric turntable according to the received output results to ensure smooth transition during starting, running and stopping.