Automatic adjustment method and system for workshop conveyor

Abstract

The present invention provides an automatic adjustment method and system for a workshop conveyor, relating to the technical field of workshop conveyor technology. The method includes real-time collection of multi-dimensional data from the workshop conveyor; feature extraction; real-time assessment of the operating status of the workshop conveyor and preliminary adjustment of the operating data; construction of a correlation network of multi-dimensional data; establishment of a multi-objective optimization model to solve the optimal adjustment strategy; and conversion of the optimal adjustment strategy into control instructions to adjust the operating data. Through intelligent data analysis and precise parameter adjustment, the present invention improves operational stability and conveying efficiency, reduces energy consumption and equipment wear, and has strong adaptability and flexibility, bringing significant economic and social benefits to modern industrial production.

Description

Technical Field

[0001] The present invention relates to the technical field of workshop conveyors, and in particular to an automatic adjustment method for a workshop conveyor and an automatic adjustment system for a workshop conveyor. Background Art

[0002] In the process of intelligent transformation of modern manufacturing industry, workshop conveyor is the core equipment for material transportation. Its operating efficiency, energy consumption control and equipment reliability directly affect the overall performance of the production line.

[0003] The invention patent with publication number CN114634005A proposes a belt conveyor drive control method and system, which can automatically obtain the driving force required by the belt conveyor; obtain the power output status of the drive motor in each drive unit of the belt conveyor according to the required driving force, and adjust the drive ratio of each drive unit according to the power output status of each drive motor. The entire process is implemented by automatic information collection and control without human participation. The characteristic of automation technology is that it can accurately record operation information, so that when the tension of the drive roller is automatically adjusted later, the actual value of the drive ratio can be automatically known, and then the tension of the drive roller can be controlled according to the actual value of the drive ratio. The entire process can be unattended.

[0004] However, existing workshop conveyor technology still has numerous flaws that are difficult to ignore. Traditional monitoring methods struggle to fully and real-timely capture the conveyor's operating status. Subtle changes in motor current and abnormal fluctuations in component temperature are often not detected in time, resulting in potential faults not being discovered and addressed promptly. Once these hidden dangers develop into failures, they can cause equipment downtime, disrupting production and resulting in significant economic losses for the company. Conveyor operating parameters are often fixed and difficult to dynamically adjust based on actual conditions. Precise adaptation to varying operating conditions is impossible, leading to problems such as unreasonable speeds and material accumulation during conveying, significantly reducing conveying efficiency. Due to the inability to accurately adjust conveyor operating parameters in real time, various components of the equipment are subjected to uneven stress and friction during long-term operation, resulting in increased wear. Frequent equipment wear not only shortens the equipment's service life, increases maintenance costs and replacement frequency, but can also impact production schedules due to equipment failures. Summary of the Invention

[0005] The present invention provides an automatic adjustment method and system for a workshop conveyor, which are used to solve the defects in the prior art.

[0006] In one aspect, the present invention provides an automatic adjustment method for a workshop conveyor, comprising:

[0007] Collect multi-dimensional data of workshop conveyors in real time, including operation data, environmental data and material data.

[0008] The wavelet transform method is used to denoise the multi-dimensional data, and principal component analysis is used to extract features to obtain operation characteristics, environmental characteristics and material characteristics.

[0009] Construct an evaluation model based on support vector machine to evaluate the operating status of workshop conveyors in real time according to operating characteristics, environmental characteristics and material characteristics, and make preliminary operating data adjustments according to preset rules based on the operating status.

[0010] The Kalman filter algorithm is used to fuse multi-dimensional data, and the long short-term memory network is used to construct the association network of multi-dimensional data.

[0011] A multi-objective optimization model of comprehensive transportation efficiency, energy consumption and equipment wear is established based on the correlation network. The multi-objective evolutionary algorithm is used to solve the multi-objective optimization model, obtain the adjustment strategy solution set, and select the optimal adjustment strategy for the current working conditions through fuzzy decision theory.

[0012] By utilizing servo drive and variable frequency speed regulation technology, the optimal adjustment strategy is converted into control instructions through a programmable logic controller, and the operating data is adjusted by combining feedforward and feedback control.

[0013] According to the present invention, an automatic adjustment method for a workshop conveyor is provided. Operational data includes operating speed, motor current, conveyor belt tension, and the rotational speed of the conveyor's rotating components. Environmental data includes conveyor component temperature and operating sound. Material data includes material weight, volume, shape, density, and distribution.

[0014] According to an automatic adjustment method for a workshop conveyor provided by the present invention, the process of extracting features using principal component analysis includes:

[0015] Construct a multidimensional data matrix, calculate the mean and standard deviation of each feature in the multidimensional data matrix, and obtain the standardized data and covariance matrix.

[0016] Perform eigenvalue decomposition on the covariance matrix, and select the eigenvectors corresponding to the first k largest eigenvalues ​​as the principal components according to the size of the eigenvalues ​​to obtain the operation characteristics, environmental characteristics and material characteristics.

[0017] According to an automatic adjustment method for a workshop conveyor provided by the present invention, the process of constructing an evaluation model based on a support vector machine includes:

[0018] Collect multi-dimensional data samples of the historical working conditions of the conveyor. Assume that there are s samples of the collected historical data, and each sample contains g features, which can be expressed as a feature matrix ,in is the feature vector of the i-th sample, and the corresponding running status label is .

[0019] Define the Gaussian kernel function, the formula is expressed as:

[0020]

[0021] Where, and represents the feature vector of the i-th sample and the j-th sample, represents the square of the Euclidean distance between the i-th sample and the j-th sample, represents the bandwidth parameter of the Gaussian kernel.

[0022] Introducing slack variables to construct the optimization problem, the formula is expressed as:

[0023]

[0024]

[0025] Where w represents the weight vector, b is the bias term, Indicates that The vector after mapping to the high-dimensional space, C represents the penalty parameter, represents the i-th running state label, represents the slack variable.

[0026] The optimization problem is converted into a dual problem, the sequential minimum optimization algorithm is used to solve the dual problem, the weight vector and bias term are calculated, and the evaluation model is obtained.

[0027] According to an automatic adjustment method for a workshop conveyor provided by the present invention, the process of using a long short-term memory network to construct a correlation network of multi-dimensional data includes:

[0028] The multi-dimensional data fused by Kalman filtering is used as the input of the long short-term memory network to obtain the input sequence.

[0029] Set up the LSTM network architecture, which includes an input layer, hidden layers, and an output layer. The input layer receives the input sequence. The hidden layer contains multiple LSTM units, which capture long-term dependencies in the input sequence. The output layer outputs the hidden state of each multi-dimensional data point in the input sequence.

[0030] The long short-term memory network is trained to analyze the time series correlation between each multi-dimensional data.

[0031] The hidden state is used as a node, and the connection weights between nodes represent the association strength between data, thus obtaining a correlation network of multi-dimensional data.

[0032] According to the present invention, an automatic adjustment method for a workshop conveyor is provided, wherein the process of establishing a multi-objective optimization model of comprehensive conveying efficiency, energy consumption, and equipment wear based on an association relationship network includes:

[0033] The conveying efficiency is defined as the ratio of the weight of the material conveyed per unit time to the maximum conveying capacity. The conveying efficiency is calculated based on the weight of the material conveyed per unit time and the maximum conveying capacity of the conveyor.

[0034] Energy consumption is defined as the electrical power consumed by the motor and is calculated based on motor power and operating time.

[0035] The wear model is established based on the conveyor belt tension variation and motor load.

[0036] A comprehensive objective function is established as a multi-objective optimization model based on transportation efficiency, energy consumption and equipment wear.

[0037] According to an automatic adjustment method for a workshop conveyor provided by the present invention, a multi-objective evolutionary algorithm is used to solve a multi-objective optimization model, and the process of obtaining an adjustment strategy solution set includes:

[0038] A certain number of individuals are randomly generated as the initial population, and each individual represents a running data vector.

[0039] For each individual, calculate its comprehensive objective function value.

[0040] The non-dominated sorting and crowding distance sorting methods are used to select outstanding individuals to enter the next generation.

[0041] Perform crossover and mutation operations on the selected individuals to generate new individuals.

[0042] The newly generated individuals are merged with the original population, and non-dominated sorting and crowding distance sorting are performed again to select a certain number of individuals to form the next generation population.

[0043] When the preset termination conditions are met, the individuals in the current population are used as the adjustment strategy solution set.

[0044] According to an automatic adjustment method for a workshop conveyor provided by the present invention, the process of selecting the optimal adjustment strategy for the current working condition by fuzzy decision theory includes:

[0045] Determining a factor set and a comment set, wherein the factors in the factor set correspond to transportation efficiency, energy consumption, and equipment wear, respectively; and the comments in the comment set represent the quality of the adjustment strategy;

[0046] For each factor, according to the performance of each strategy in the adjustment strategy solution under this factor, its membership function is determined to obtain the fuzzy relationship matrix;

[0047] Determine the weight vector of each factor based on the requirements of the current working conditions;

[0048] A fuzzy comprehensive evaluation vector is calculated, and according to the value of each element in the fuzzy comprehensive evaluation vector, a regulation strategy corresponding to the element with the largest membership degree is selected as the optimal regulation strategy for the current working condition.

[0049] According to the present invention, an automatic adjustment method for a workshop conveyor is provided, wherein the process of adjusting operating data by combining feedforward and feedback control includes:

[0050] The optimal adjustment strategy is received through a programmable logic controller, and key adjustment parameters in the optimal adjustment strategy are extracted.

[0051] A feedforward control model is established based on the physical characteristics and historical operating data of the workshop conveyor.

[0052] The feedforward control model is used to calculate the corresponding feedforward control instructions based on the key adjustment parameters.

[0053] The feedforward control instructions are sent to the servo drive and variable frequency speed regulation equipment to adjust the operating data of the conveyor.

[0054] On the other hand, the present invention also provides an automatic adjustment system for a workshop conveyor, comprising:

[0055] The data acquisition module is used to collect multi-dimensional data of workshop conveyors in real time. The multi-dimensional data includes operation data, environmental data and material data.

[0056] The feature extraction module is used to use the wavelet transform method to remove noise from multiple dimensions and use principal component analysis to extract features to obtain operation features, environmental features and material features.

[0057] The preliminary adjustment module is used to build an evaluation model based on support vector machines. It evaluates the operating status of the workshop conveyor in real time according to the operating characteristics, environmental characteristics and material characteristics, and makes preliminary operating data adjustments according to the preset rules based on the operating status.

[0058] The association relationship building module is used to fuse multi-dimensional data using the Kalman filter algorithm and build an association relationship network for multi-dimensional data using the long short-term memory network.

[0059] The regulation strategy generation module is used to establish a multi-objective optimization model of comprehensive transportation efficiency, energy consumption, and equipment wear based on the association relationship network, adopt a multi-objective evolutionary algorithm to solve the multi-objective optimization model, obtain the regulation strategy solution set, and select the optimal regulation strategy for the current working conditions through fuzzy decision theory.

[0060] The control and regulation module is used to utilize servo drive and variable frequency speed regulation technology to convert the optimal regulation strategy into control instructions through a programmable logic controller, and to adjust the operating data in combination with feedforward and feedback control.

[0061] The present invention provides an automatic adjustment method and system for a workshop conveyor. By collecting multi-dimensional data in real time and performing denoising, feature extraction, and status evaluation, it is possible to promptly detect abnormal conditions during the operation of the conveyor and make preliminary operating data adjustments according to preset rules to avoid further deterioration of the fault, thereby ensuring the continuous and stable operation of the conveyor and reducing production interruptions caused by equipment failure.

[0062] The data association network constructed through the long-short-term memory network can deeply explore the inherent connections between multi-dimensional data. In combination with multi-objective optimization models and fuzzy decision-making theory, it can dynamically adjust conveyor operating parameters such as operating speed and conveyor belt tension based on current material characteristics and environmental conditions. This allows the conveyor to operate in an optimal state, quickly and efficiently completing material transportation tasks, and significantly improving production efficiency.

[0063] By comprehensively considering conveying efficiency, energy consumption, and equipment wear, a multi-objective optimization model can minimize conveyor energy consumption while maintaining conveying efficiency. Precisely adjusting operating parameters avoids unnecessary energy waste and achieves a balance between energy conservation and efficiency. Furthermore, proper adjustment of operating parameters can reduce equipment wear, extend its service life, and reduce maintenance costs and replacement frequency. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0065] Figure 1 This is a flow chart of an automatic adjustment method for a workshop conveyor provided by an embodiment of the present invention;

[0066] Figure 2 It is a structural schematic diagram of an automatic adjustment system for a workshop conveyor provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0067] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0068] The following combination Figure 1-Figure 2 The present invention describes an automatic adjustment method and system for a workshop conveyor.

[0069] Figure 1 It is a flow chart of an automatic adjustment method for a workshop conveyor provided by an embodiment of the present invention.

[0070] like Figure 1 As shown, an embodiment of the present invention provides an automatic adjustment method and system for a workshop conveyor. The execution subject may be an automatic adjustment method for a workshop conveyor, and the method includes:

[0071] Collect multi-dimensional data of workshop conveyors in real time, including operation data, environmental data and material data.

[0072] Operational data includes operating speed, motor current, belt tension, and the speed of the conveyor's rotating components. Environmental data includes conveyor component temperature and operating sound. Material data includes material weight, volume, shape, density, and distribution.

[0073] The running speed is collected using a speed sensor. The speed sensor is installed at a specific position on the conveyor's drive shaft or conveyor belt to measure its linear speed or angular speed by sensing the rotation of the shaft or the movement of the conveyor belt.

[0074] The motor current is measured using a current transformer. Installed in the motor's power supply circuit, the current transformer converts the high current into a proportionally small current signal, which is then converted into a collectible electrical signal by a signal processing circuit.

[0075] Use a tension sensor to measure conveyor belt tension. Install the tension sensor on the conveyor belt's support roller or tensioning device and determine the conveyor belt tension by detecting the force applied by the conveyor belt to the sensor.

[0076] The speed of the conveyor's rotating components is measured using speed sensors, such as magnetoelectric or photoelectric sensors. These sensors are installed near rotating components (such as rollers and gears) and detect the component's rotational frequency to obtain speed information.

[0077] Use temperature sensors, such as thermocouples or RTDs. These sensors can be installed on the surface of key conveyor components (such as motors, bearings, etc.) to monitor component temperature changes in real time.

[0078] Use sound sensors to collect operating sounds. The sound sensors are installed at appropriate locations around the conveyor and can capture the sound signals generated during the conveyor's operation and convert them into electrical signals.

[0079] Install a load cell at the feed port or a specific location on the conveyor. When the material passes through the load cell, the sensor measures the weight of the material.

[0080] Measure the volume of the material using a volumetric device, such as a laser scanner or ultrasonic sensor. These devices can be mounted above the conveyor and calculate the volume by scanning the outline of the material.

[0081] Use an industrial camera to capture the shape of the material. Install the industrial camera above or on the side of the conveyor to take pictures of the material. Then use image processing technology to analyze the captured images and identify the shape characteristics of the material.

[0082] Based on the collected material weight and volume data, the density of the material is calculated using the density calculation formula.

[0083] Combine industrial cameras and image processing technology, or use multiple sensor arrays to detect the distribution position and density of materials on the conveyor belt and collect material distribution data.

[0084] By collecting operational data such as operating speed, motor current, belt tension, and the speed of rotating conveyor components, the conveyor's operating status can be monitored in real time. For example, abnormal changes in motor current may indicate excessive motor load or a fault; unstable belt tension may cause the belt to slip or deviate. Combined with environmental data such as component temperature and operating sound, it is possible to further determine whether the conveyor is operating normally. For example, excessively high component temperature may indicate wear or poor heat dissipation; abnormal operating sound may indicate a mechanical failure. By extracting features from this data and inputting it into an evaluation model, the conveyor's operating status can be accurately assessed, providing a basis for subsequent operational adjustments.

[0085] Material data, including weight, volume, shape, density, and distribution, is crucial for optimizing conveyor operations. Materials of varying weights and volumes place varying demands on the conveyor's load. These characteristics allow for adjustments to conveyor speed and motor power to improve efficiency and reduce energy consumption. Material shape and distribution can affect conveyor belt force uniformity and operational stability. Collecting this data allows for adjustments to minimize wear and extend equipment life.

[0086] Continuously collecting multi-dimensional data can also help establish a historical database of conveyor operations. By analyzing and mining this historical data, we can identify trends and patterns in the data. When unusual fluctuations in certain data occur, we can predict potential equipment failures in advance, enabling timely maintenance measures to avoid production interruptions and financial losses caused by equipment failures.

[0087] The wavelet transform method is used to denoise the multi-dimensional data, and principal component analysis is used to extract features to obtain operation characteristics, environmental characteristics and material characteristics.

[0088] The process of denoising multidimensional data using wavelet transform method includes:

[0089] The multi-dimensional data of the workshop conveyor collected in real time is stored in the form of a matrix or vector, with the data of each dimension as a column of the matrix or an element of the vector.

[0090] According to the characteristics of multi-dimensional data and denoising requirements, select appropriate wavelet basis functions, such as Haar wavelet, Daubechies wavelet, etc.

[0091] Determine the number of wavelet decomposition layers. The number of decomposition layers will affect the denoising effect and needs to be reasonably selected based on the characteristics of the data and the noise level.

[0092] Wavelet decomposition is performed on the organized multidimensional data. Specifically, the original data is decomposed into wavelet coefficients of different scales and frequencies using a wavelet transform. At each decomposition level, the data is divided into approximate coefficients (low-frequency components) and detail coefficients (high-frequency components). Approximate coefficients represent the primary characteristics of the data, while detail coefficients incorporate noise and high-frequency information. Through multiple decompositions, approximate and detail coefficients at different scales can be obtained.

[0093] The detail coefficient is compared with a selected threshold and processed based on the comparison result. Generally, there are two methods of processing: hard thresholding and soft thresholding. Hard thresholding sets detail coefficients with absolute values ​​less than the threshold to zero, while leaving them unchanged. Soft thresholding sets detail coefficients with absolute values ​​less than the threshold to zero, while leaving them with absolute values ​​greater than the threshold less than the threshold.

[0094] After thresholding the detail coefficients, the processed detail coefficients and the unprocessed approximate coefficients are reconstructed using wavelet transforms. Through inverse wavelet transforms, the approximate coefficients at different scales and the processed detail coefficients are combined to obtain denoised multi-dimensional data.

[0095] The process of feature extraction using principal component analysis includes:

[0096] Construct a multi-dimensional data matrix, the formula is expressed as:

[0097]

[0098] In the formula, n represents the number of samples, m represents the number of features, represents the jth eigenvalue of the i-th sample.

[0099] For each feature, calculate its mean and standard deviation to obtain the standardized data and covariance matrix, which is expressed as follows:

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] Where, represents the mean of the jth feature, represents the standard deviation of the j-th feature, represents the standardized data of the jth feature of the i-th sample, Z represents the standardized data matrix, and S represents the covariance matrix.

[0106] The eigenvalue decomposition of the covariance matrix is ​​expressed as follows:

[0107]

[0108] Where P represents an orthogonal matrix composed of eigenvectors, represents a diagonal matrix consisting of eigenvalues.

[0109] According to the size of the eigenvalue, the eigenvectors corresponding to the first k largest eigenvalues ​​are selected as principal components, and the principal component scores are calculated to obtain the operation characteristics, environmental characteristics and material characteristics.

[0110] The formula of the principal component score matrix is ​​expressed as:

[0111]

[0112] Where, It represents an m×k matrix composed of the first k eigenvectors, and the principal component scores are the extracted operating characteristics, environmental characteristics, and material characteristics.

[0113] Construct an evaluation model based on support vector machine to evaluate the operating status of workshop conveyors in real time according to operating characteristics, environmental characteristics and material characteristics, and make preliminary operating data adjustments according to preset rules based on the operating status.

[0114] The process of building an evaluation model based on support vector machines includes:

[0115] Collect multi-dimensional data samples of the historical working conditions of the conveyor. Assume that there are s samples of the collected historical data, and each sample contains g features, which can be expressed as a feature matrix ,in is the feature vector of the i-th sample, and the corresponding running status label is .

[0116] Define the Gaussian kernel function, the formula is expressed as:

[0117]

[0118] Where, and represents the feature vector of the i-th sample and the j-th sample, represents the square of the Euclidean distance between the i-th sample and the j-th sample, represents the bandwidth parameter of the Gaussian kernel.

[0119] Introducing slack variables to construct the optimization problem, the formula is expressed as:

[0120]

[0121]

[0122] Where w represents the weight vector, b is the bias term, Indicates that The vector after mapping to the high-dimensional space, C represents the penalty parameter, represents the i-th running state label, represents the slack variable.

[0123] The optimization problem is converted into a dual problem, which is expressed as follows:

[0124]

[0125]

[0126] Where, represents the Lagrange multiplier vector.

[0127] Use the sequential minimal optimization algorithm to solve the dual problem and continuously update and until the preset convergence condition is met.

[0128] After solving the optimal Lagrange multiplier, calculate the weight vector, which is expressed as:

[0129]

[0130] Where, represents the optimal Lagrange multiplier.

[0131] Select one The support vector of , calculate the bias term, the formula is expressed as:

[0132]

[0133] Where, represents the kth running state label, represents the kth support vector.

[0134] The obtained support vector machine decision function is used to obtain the evaluation model. The support vector machine decision function formula is expressed as:

[0135]

[0136] Where sgn(z) represents the sign function, when z ≥ 0, sgn(z) = 1, when z < 0, sgn(z) = -1, and q represents a new sample.

[0137] The process of adjusting preliminary operating data according to preset rules based on the operating status includes:

[0138] The operating status is divided into normal, slightly abnormal, moderately abnormal and severely abnormal categories.

[0139] When the running status is normal, the current running data remains unchanged.

[0140] When the operating status is slightly abnormal, such as the operating speed is slightly lower than the normal range, increase the operating speed according to the preset ratio. If the conveyor belt tension is slightly low, appropriately increase the pressure of the tension adjustment device.

[0141] When the operating status is moderately abnormal, increase the adjustment efforts, such as increasing the operating speed by 10%, and monitor and fine-tune parameters such as motor current.

[0142] When the operating status is seriously abnormal, the conveyor will be stopped immediately, an alarm will be sounded, and troubleshooting and repairs will be carried out.

[0143] The Kalman filter algorithm is used to fuse multi-dimensional data, and the long short-term memory network is used to construct the association network of multi-dimensional data.

[0144] The process of fusing multi-dimensional data using the Kalman filter algorithm includes:

[0145] By comprehensively considering multiple dimensions of information, including operational data, environmental data, and material data, a state vector is determined that can describe the operating status of the workshop conveyor. For example, the state vector can include key parameters such as the conveyor belt speed and tension, and the motor current.

[0146] Based on the physical characteristics and operating rules of the conveyor, a state transition equation is established to describe the relationship between the system state and time. The state transition equation is expressed as a linear or nonlinear mathematical model to reflect how the state at one moment evolves to the state at the next moment.

[0147] Determine observable multi-dimensional data, such as operating speed, motor current, component temperature, and other data collected by various sensors.

[0148] Establish an observation equation to describe the relationship between the observation vector and the system state vector. This equation shows how to obtain the actual observed data from the system state.

[0149] Based on the prior knowledge or initial measurement value of the system, the initial state of the system is estimated to obtain the estimated value of the initial state vector.

[0150] Determine the covariance matrix of the initial state estimate, which reflects the uncertainty of the initial state estimate.

[0151] According to the state transition equation, the state estimate at the previous moment is used to predict the system state at the current moment.

[0152] According to the state transfer equation and the initial covariance matrix, the covariance matrix of the current state estimate is predicted.

[0153] The Kalman gain is calculated based on the prediction covariance matrix, the observation equation, and the observation noise covariance matrix. The Kalman gain is used to weigh the weight of the predicted value and the observed value in the state update.

[0154] Combining the observed value and predicted state at the current moment, the Kalman gain is used to update the system state to obtain the optimal state estimate at the current moment.

[0155] According to the Kalman gain and the predicted covariance matrix, the covariance matrix of the state estimate is updated to reflect the uncertainty of the updated state estimate.

[0156] As new observations continue to arrive, predictions and updates are continuously performed, enabling real-time fusion of multi-dimensional data. Through continuous iteration, the Kalman filter algorithm can adaptively adjust state estimates, reducing the impact of noise and uncertainty, and improving the accuracy and reliability of data fusion.

[0157] The system state estimate after fusion by the Kalman filter algorithm is output as the final fusion result. This result integrates information from multi-dimensional data and can more accurately reflect the operating status of the workshop conveyor.

[0158] The process of using long short-term memory networks to construct a multi-dimensional data association network includes:

[0159] The multi-dimensional data after Kalman filter fusion is used as the input of the long short-term memory network. Suppose the input sequence is x1, x2, ..., x T , where x t Represents a multi-dimensional data vector at time t.

[0160] Set up the LSTM network architecture, which includes an input layer, hidden layers, and an output layer. The input layer receives the input sequence. The hidden layer contains multiple LSTM units, which capture long-term dependencies in the input sequence. The output layer outputs the hidden state of each multi-dimensional data point in the input sequence.

[0161] The long short-term memory network is trained to analyze the time series correlation between each multi-dimensional data.

[0162] The hidden state is used as a node, and the connection weights between nodes represent the association strength between data, thus obtaining a correlation network of multi-dimensional data.

[0163] A multi-objective optimization model of comprehensive transportation efficiency, energy consumption and equipment wear is established based on the correlation network. The multi-objective evolutionary algorithm is used to solve the multi-objective optimization model, obtain the adjustment strategy solution set, and select the optimal adjustment strategy for the current working conditions through fuzzy decision theory.

[0164] The process of establishing a multi-objective optimization model for comprehensive transportation efficiency, energy consumption, and equipment wear based on the association relationship network includes:

[0165] The conveying efficiency is defined as the ratio of the weight of the material conveyed per unit time to the maximum conveying capacity. The conveying efficiency is calculated based on the weight of the material conveyed per unit time and the maximum conveying capacity of the conveyor. The formula is:

[0166]

[0167] Where, Indicates the transport efficiency, It indicates the weight of material transported per unit time, and M indicates the maximum transport capacity.

[0168] Energy consumption is defined as the power consumption of the motor. The energy consumption is calculated based on the motor power and running time. The formula is expressed as:

[0169]

[0170] Where, represents energy consumption, P represents motor power, and t represents running time.

[0171] The wear model is established based on the tension change of the conveyor belt and the motor load. The formula is expressed as follows:

[0172]

[0173] Where, Indicates equipment wear, Indicates the change in conveyor belt tension. Indicates the change in motor current, and Represents the weight coefficient.

[0174] A comprehensive objective function is established as a multi-objective optimization model based on transportation efficiency, energy consumption, and equipment wear. The formula of the comprehensive objective function is expressed as follows:

[0175]

[0176] Where x represents the running data vector.

[0177] The process of using a multi-objective evolutionary algorithm to solve a multi-objective optimization model and obtain the adjustment strategy solution set includes:

[0178] A certain number of individuals are randomly generated as the initial population, and each individual represents a running data vector.

[0179] For each individual, calculate its comprehensive objective function value F(x).

[0180] The non-dominated sorting and crowding distance sorting methods are used to select outstanding individuals to enter the next generation.

[0181] Perform crossover and mutation operations on the selected individuals to generate new individuals.

[0182] The newly generated individuals are merged with the original population, and non-dominated sorting and crowding distance sorting are performed again to select a certain number of individuals to form the next generation population.

[0183] When the preset termination conditions are met, the individuals in the current population are used as the adjustment strategy solution set.

[0184] The process of selecting the optimal regulation strategy for the current working conditions through fuzzy decision theory includes:

[0185] Determine the factor set and comment set, factor set U = {u1,u2,u3}, u1, u2 and u3 correspond to transportation efficiency, energy consumption and equipment wear respectively. Comment set V = {v1,v2,…,v n}, v i Indicates the quality of the adjustment strategy.

[0186] For each factor u i According to the performance of each strategy in the adjustment strategy solution under this factor, its membership function is determined and the fuzzy relationship matrix is ​​obtained. , where r ij It represents the membership of the i-th factor to the j-th strategy.

[0187] Determine the weight vector of each factor according to the requirements of the current working conditions ,in, and , represents the i-th weight vector.

[0188] Calculate the fuzzy comprehensive evaluation vector B=A·R, where · represents the fuzzy synthesis operator.

[0189] According to the value of each element in the fuzzy comprehensive evaluation vector B, the regulation strategy corresponding to the element with the largest membership degree is selected as the optimal regulation strategy for the current working condition.

[0190] Utilizing servo drive and variable frequency speed regulation technology, the optimal regulation strategy is converted into control instructions through a programmable logic controller, and operating data is adjusted by combining feedforward and feedback control. The process includes:

[0191] The optimal adjustment strategy is received through a programmable logic controller, and key adjustment parameters in the optimal adjustment strategy are extracted.

[0192] The programmable logic controller receives the optimal adjustment strategy for the current operating conditions from the system, which is selected using fuzzy decision theory. This strategy includes specific adjustment parameters and targets based on the workshop conveyor's operating data, such as conveying speed, power, and material delivery volume.

[0193] The PLC analyzes the optimal control strategy and extracts key control parameters and target values. For example, if the control strategy requires increasing the conveyor speed from the current 50 m / min to 60 m / min, the PLC can identify the speed adjustment information.

[0194] A feedforward control model was developed based on the physical characteristics and historical operating data of the shop floor conveyor. This model describes the relationship between input variables (such as parameters in the regulation strategy) and output variables (such as the conveyor's operating status). For example, through experiments and data analysis, the functional relationship between conveyor speed and motor input voltage was determined.

[0195] Using the feedforward control model, we calculate the corresponding feedforward control instructions based on the optimal adjustment strategy parameters obtained analytically. Taking the aforementioned speed adjustment as an example, we calculate the motor input voltage required to increase the conveyor speed to 60 m / min based on the functional relationship between speed and motor input voltage.

[0196] The programmable logic controller sends the calculated feedforward control instructions to the servo drive and variable frequency speed regulation equipment to adjust the operation of the conveyor in advance, reducing response time and errors.

[0197] In summary, this embodiment provides an automatic adjustment method for a workshop conveyor. By collecting multi-dimensional data in real time and performing denoising, feature extraction, and status assessment, anomalies during conveyor operation can be promptly detected. For example, sudden changes in motor current and abnormal increases in component temperature can be accurately captured. Once an anomaly is detected, the system will quickly make preliminary operating data adjustments according to preset rules, promptly correcting any potential problems and preventing further deterioration of the fault, thereby ensuring the continuous and stable operation of the conveyor and reducing production interruptions caused by equipment failure.

[0198] The data association network constructed through the long-short-term memory network can deeply explore the inherent connections between multi-dimensional data. In combination with multi-objective optimization models and fuzzy decision-making theory, it can dynamically adjust conveyor operating parameters such as operating speed and conveyor belt tension based on current material characteristics and environmental conditions. This allows the conveyor to operate in an optimal state, quickly and efficiently completing material transportation tasks, and significantly improving production efficiency.

[0199] By comprehensively considering conveying efficiency, energy consumption, and equipment wear, a multi-objective optimization model can minimize conveyor energy consumption while maintaining conveying efficiency. Precisely adjusting operating parameters avoids unnecessary energy waste and achieves a balance between energy conservation and efficiency. Furthermore, proper adjustment of operating parameters can reduce equipment wear, extend its service life, and reduce maintenance costs and replacement frequency.

[0200] Based on the same general inventive concept, the present invention also protects an automatic adjustment system for a workshop conveyor. The automatic adjustment system for a workshop conveyor provided by the present invention is described below. The automatic adjustment system for a workshop conveyor described below and the automatic adjustment method for a workshop conveyor described above can be referenced to each other.

[0201] Figure 2 It is a structural schematic diagram of an automatic adjustment system for a workshop conveyor provided by an embodiment of the present invention.

[0202] like Figure 2 As shown, an automatic adjustment system for a workshop conveyor includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor includes a data acquisition module, a feature extraction module, a preliminary adjustment module, an association relationship building module, an adjustment strategy generation module, and a control and adjustment module.

[0203] The data acquisition module is used to collect multi-dimensional data of workshop conveyors in real time. The multi-dimensional data includes operation data, environmental data and material data.

[0204] The feature extraction module is used to use the wavelet transform method to remove noise from multiple dimensions and use principal component analysis to extract features to obtain operation features, environmental features and material features.

[0205] The preliminary adjustment module is used to build an evaluation model based on support vector machines. It evaluates the operating status of the workshop conveyor in real time according to the operating characteristics, environmental characteristics and material characteristics, and makes preliminary operating data adjustments according to the preset rules based on the operating status.

[0206] The association relationship building module is used to fuse multi-dimensional data using the Kalman filter algorithm and to build an association relationship network for multi-dimensional data using the long short-term memory network.

[0207] The regulation strategy generation module is used to establish a multi-objective optimization model of comprehensive transportation efficiency, energy consumption, and equipment wear based on the association relationship network, adopt a multi-objective evolutionary algorithm to solve the multi-objective optimization model, obtain the regulation strategy solution set, and select the optimal regulation strategy for the current working conditions through fuzzy decision theory.

[0208] The control and regulation module is used to utilize servo drive and variable frequency speed regulation technology to convert the optimal regulation strategy into control instructions through a programmable logic controller, and to regulate operating data in combination with feedforward and feedback control.

[0209] Through the description of the above embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0210] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An automatic adjustment method for a workshop conveyor, characterized in that: include: Collect multi-dimensional data of workshop conveyors in real time, including operation data, environmental data and material data; Using wavelet transform method to remove noise from the multi-dimensional data, and using principal component analysis to extract features to obtain operation features, environmental features and material features; Constructing an evaluation model based on a support vector machine to evaluate the operating status of the workshop conveyor in real time according to the operating characteristics, environmental characteristics, and material characteristics, and making preliminary operating data adjustments according to preset rules based on the operating status; Using a Kalman filter algorithm to fuse the multi-dimensional data, and using a long short-term memory network to construct a correlation network of the multi-dimensional data; Establishing a multi-objective optimization model for comprehensive transportation efficiency, energy consumption, and equipment wear based on the association relationship network, solving the multi-objective optimization model using a multi-objective evolutionary algorithm to obtain a set of adjustment strategy solutions, and selecting the optimal adjustment strategy for the current working conditions using fuzzy decision theory; By utilizing servo drive and variable frequency speed regulation technology, the optimal regulation strategy is converted into control instructions through a programmable logic controller, and the operating data is regulated by combining feedforward and feedback control.

2. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The operating data includes operating speed, motor current, conveyor belt tension and rotation speed of conveyor rotating parts; the environmental data includes conveyor component temperature and operating sound; the material data includes material weight, volume, shape, density and distribution.

3. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of feature extraction using principal component analysis includes: Constructing a multidimensional data matrix, and calculating the mean and standard deviation of each feature in the multidimensional data matrix to obtain standardized data and a covariance matrix; Perform eigenvalue decomposition on the covariance matrix, and select the eigenvectors corresponding to the first k largest eigenvalues ​​as the principal components according to the size of the eigenvalues ​​to obtain the operation characteristics, environmental characteristics and material characteristics.

4. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of building an evaluation model based on support vector machines includes: Collect multi-dimensional data samples of the historical working conditions of the conveyor. Assume that there are s samples of the collected historical data, and each sample contains g features, which can be expressed as a feature matrix ,in is the feature vector of the i-th sample, and the corresponding running status label is ; Define the Gaussian kernel function, the formula is expressed as: Where, and represents the feature vector of the i-th sample and the j-th sample, represents the square of the Euclidean distance between the i-th sample and the j-th sample, represents the bandwidth parameter of the Gaussian kernel; Introducing slack variables to construct the optimization problem, the formula is expressed as: Where w represents the weight vector, b is the bias term, Indicates that The vector after mapping to the high-dimensional space, C represents the penalty parameter, represents the i-th running state label, represents the slack variable; The optimization problem is converted into a dual problem, the dual problem is solved using a sequential minimum optimization algorithm, and the weight vector and the bias term are calculated to obtain an evaluation model.

5. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of constructing the association relationship network of the multi-dimensional data using the long short-term memory network includes: The multi-dimensional data after Kalman filter fusion is used as the input of the long short-term memory network to obtain the input sequence; The structure of the long short-term memory network is set to include an input layer, a hidden layer, and an output layer; the input layer is used to receive the input sequence; the hidden layer includes multiple long short-term memory network units, which are used to capture long-term dependencies in the input sequence; and the output layer is used to output the hidden state of each multi-dimensional data in the input sequence; Training the long short-term memory network to analyze the time series correlation between each of the multi-dimensional data; The hidden states are used as nodes, and the connection weights between the nodes represent the association strength between the data, thereby obtaining an association relationship network of multi-dimensional data.

6. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of establishing a multi-objective optimization model for comprehensive transportation efficiency, energy consumption, and equipment wear based on the association relationship network includes: The conveying efficiency is defined as the ratio of the weight of the material conveyed per unit time to the maximum conveying capacity. The conveying efficiency is calculated based on the weight of the material conveyed per unit time and the maximum conveying capacity of the conveyor. Energy consumption is defined as the power consumption of the motor, and is calculated based on motor power and operating time; Establish a wear model based on the tension change of the conveyor belt and the motor load; A comprehensive objective function is established as a multi-objective optimization model based on the transportation efficiency, energy consumption and equipment wear.

7. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of using a multi-objective evolutionary algorithm to solve the multi-objective optimization model and obtain a set of adjustment strategy solutions includes: Randomly generate a certain number of individuals as an initial population, wherein the individuals represent running data vectors; For each individual, calculate its comprehensive objective function value; Use non-dominated sorting and crowding distance sorting methods to select outstanding individuals to enter the next generation; Perform crossover and mutation operations on the selected individuals to generate new individuals; Merge the newly generated individuals with the original population, perform non-dominated sorting and crowding distance sorting again, and select a certain number of individuals to form the next generation population; When the preset termination conditions are met, the individuals in the current population are used as the adjustment strategy solution set.

8. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of selecting the optimal regulation strategy for the current working conditions through fuzzy decision theory includes: Determining a factor set and a comment set, wherein the factors in the factor set correspond to transportation efficiency, energy consumption, and equipment wear, respectively; and the comments in the comment set represent the quality of the adjustment strategy; For each factor, according to the performance of each strategy in the adjustment strategy solution under this factor, its membership function is determined to obtain the fuzzy relationship matrix; Determine the weight vector of each factor based on the requirements of the current working conditions; A fuzzy comprehensive evaluation vector is calculated, and according to the value of each element in the fuzzy comprehensive evaluation vector, a regulation strategy corresponding to the element with the largest membership degree is selected as the optimal regulation strategy for the current working condition.

9. The automatic adjustment method for a workshop conveyor according to claim 1, characterized in that: The process of adjusting operating data by combining feedforward and feedback control includes: Receiving the optimal adjustment strategy through a programmable logic controller and extracting key adjustment parameters in the optimal adjustment strategy; Establish a feedforward control model based on the physical characteristics and historical operating data of the workshop conveyor; Utilizing the feedforward control model, and according to the key adjustment parameters, calculating corresponding feedforward control instructions; The feedforward control instruction is sent to the servo drive and frequency conversion speed regulation equipment to adjust the operation data of the conveyor.

10. An automatic adjustment system for a workshop conveyor, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, an automatic adjustment method for a workshop conveyor as described in any one of claims 1 to 9 is implemented.

Images