Industrial production control system based on industrial internet of things

Abstract

The invention relates to the technical field of industrial production, and discloses an industrial production control system based on industrial Internet of Things, which comprises a data acquisition module, a data preprocessing module, a real-time monitoring module, a production prediction module, a comparison module, a production optimization module, an adjustment control module and a man-machine interaction module. The production prediction module and the production optimization module are arranged, so that the yield of the production line can be predicted based on the constructed neural network model, and the predicted production yield can be obtained; if the production demand can be met, the production scheme is further optimized, and the optimal production scheme is obtained based on a genetic algorithm, so that the production cost is minimized, the intelligence of industrial production is improved, and the industrial production cost is reduced; and meanwhile, a real-time monitoring module is arranged to monitor key parameters such as temperature and pressure of production equipment in real time, so that the situation that the production equipment breaks down, a production line cannot operate, and the industrial production efficiency is reduced is prevented.

Description

Technical Field

[0001] The present invention relates to the field of industrial production technology, and more specifically to an industrial production control system based on industrial Internet of Things. Background Art

[0002] In recent years, the industrial field has ushered in a wave of intelligence and automation, which has not only improved production efficiency but also optimized resource utilization. The Industrial Internet of Things is sensors and instruments connected to industrial applications through the Internet. The network uses IoT sensing and communication technologies to collect industrial application data and analyze the data. The purpose is to optimize the production process, improve production efficiency, reduce manufacturing costs, and improve product quality, and ultimately achieve the upgrading of traditional industries to a new stage of intelligence; production lines and production workshops, as the core places of production, undertake various production tasks and require a high degree of production automation and control systems to support and manage.

[0003] However, most traditional industrial production control systems only control basic parameters such as temperature and pressure of production lines and production equipment. They are unable to analyze actual production needs and predicted production conditions, and adjust the optimal productivity of production lines. They are unable to ensure that industrial production needs are met under the conditions required, and that industrial production is carried out through the optimal production plan, and are unable to improve the level of intelligence in industrial production. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an industrial production control system based on industrial Internet of Things to solve the problems existing in the above-mentioned background technology.

[0005] The present invention provides the following technical solutions: an industrial production control system based on the industrial Internet of Things, comprising a data acquisition module, a data preprocessing module, a real-time monitoring module, a production prediction module, a comparison module, a production optimization module, an adjustment control module and a human-computer interaction module;

[0006] The data acquisition module is used to collect production equipment data and transmit it to the data preprocessing module;

[0007] The data preprocessing module is used to preprocess the data collected by the data acquisition module to obtain production equipment data that can be directly used, and transmit it to the real-time monitoring module and the production prediction module;

[0008] The real-time monitoring module is used to receive data from the data preprocessing module, monitor the temperature and pressure of the production equipment in real time, and send an early warning notification to the human-computer interaction module if the set threshold is exceeded;

[0009] The production prediction module predicts the output of the production line based on the constructed neural network model to obtain the predicted production output;

[0010] The comparison module compares the predicted production output of the production prediction module with the production demand. If the production demand is reached, an optimization instruction is sent to the production optimization module; otherwise, an early warning prompt is sent to the human-computer interaction module.

[0011] The production optimization module receives the optimization instruction sent by the comparison module, generates a production plan, obtains the optimal production plan based on the genetic algorithm, and sends a control instruction to the adjustment control module;

[0012] The adjustment control module is used to receive data from the real-time monitoring module and control instructions from the production optimization module to adjust the production equipment accordingly;

[0013] The human-computer interaction module is used to perform human-computer interaction display on the data.

[0014] Preferably, the neural network model is an LSTM neural network model, and the construction method includes:

[0015] Initialize the basic architecture of the LSTM neural network model, which includes an input layer, two LSTM units, a Dropout layer, a fully connected layer, a learnable parameter activation function, a comprehensive loss function, an optimizer, a regularizer, and an output layer; the output layer includes n nodes;

[0016] Each LSTM unit consists of a memory unit and three control gates; the three control gates are input gate, forget gate and output gate; a fully connected layer is added between LSTM units;

[0017] The n output layers are used to output the production output of the n production lines.

[0018] Preferably, the learnable parameter activation function is: Where s is the input data; p1 is a learnable slope parameter that controls the slope of the linear transformation of s; p2 is a learnable offset parameter that controls the offset of the Sigmoid function;

[0019] Uncertainty modeling is introduced into the loss function to obtain the comprehensive loss function Loss:

[0020]

[0021] ; Among them, q is the quantile, which represents the quantile level in the loss calculation; q takes a value between (0, 1); y true is the real target value, indicating the actual production line output; is the lower quantile of the predicted distribution; is the upper quantile of the predicted distribution; | | + is a positive function;

[0022] The optimizer adopts an adaptive optimizer that adaptively adjusts the learning rate.

[0023] Preferably, the optimization formula of the adaptive optimizer is: in, is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ+1; is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ; η l is the learning rate of the lth layer of the LSTM neural network model, and different learning rates are selected according to different layers; v λ is the exponential moving average of the squared gradients for training iterations λ; θ is a small constant to prevent the denominator from being zero; m λ，l is the parameter gradient for the training iteration number λ.

[0024] Preferably, in the loss function, if but in is the value of the comprehensive loss function when the number of training iterations is λ; is the value of the comprehensive loss function when the number of training iterations is λ-1; A is the reduction parameter.

[0025] Preferably, the objective function of the production optimization module to generate the production plan is: Among them, Z′ is the production demand, that is, the production output to be achieved; C ij is the output of the i-th production line at the j-th moment, j = 1, 2, 3...t, i = 1, 2, 3...n.

[0026] Preferably, the genetic algorithm to obtain the optimal production plan includes the following steps:

[0027] Step S01: Encode the production plan, obtain chromosomes, and construct the initial population;

[0028] Step S02: Determine the fitness function;

[0029] Step S03: Perform natural selection on chromosomes in the population;

[0030] Step S04: performing crossover recombination on chromosomes in the population;

[0031] Step S05: mutating the chromosomes in the population;

[0032] Step S06: Obtain a new population, preset the population generation to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps S03 to S05 until the generation corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is taken as the optimal production plan.

[0033] Technical effects and advantages of the present invention:

[0034] The present invention is provided with a production prediction module and a production optimization module, which is conducive to predicting the output of the production line based on the constructed neural network model and obtaining the predicted production output; if the production demand can be met, the production plan is further optimized, and the optimal production plan is obtained based on the genetic algorithm, so as to minimize the production cost, improve the intelligence of industrial production, and reduce the industrial production cost; at the same time, a real-time monitoring module is provided to monitor the key parameters such as the temperature and pressure of the production equipment in real time to prevent the production equipment from malfunctioning and causing the production line to be unable to operate, thereby reducing the industrial production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a structural diagram of the industrial production control system based on the industrial Internet of Things of the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The industrial production control system based on industrial Internet of Things involved in the present invention is not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0037] The present invention provides an industrial production control system based on the industrial Internet of Things, including a data acquisition module, a data preprocessing module, a real-time monitoring module, a production prediction module, a comparison module, a production optimization module, an adjustment control module and a human-computer interaction module;

[0038] The data acquisition module is used to collect production equipment data and transmit it to the data preprocessing module; the production equipment data includes but is not limited to production equipment temperature, pressure, working time, output and equipment energy consumption;

[0039] The data preprocessing module is used to preprocess the data collected by the data acquisition module to obtain production equipment data that can be directly used, and transmit it to the real-time monitoring module and the production prediction module;

[0040] The real-time monitoring module is used to receive data from the data preprocessing module, and to monitor key parameters such as temperature and pressure of the production equipment in real time. If the set threshold is exceeded, an early warning notification is sent to the human-computer interaction module;

[0041] The production prediction module predicts the output of the production line based on the constructed neural network model to obtain the predicted production output;

[0042] The comparison module compares the predicted production output of the production prediction module with the production demand. If the production demand can be met, an optimization instruction is sent to the production optimization module; otherwise, an early warning prompt is sent to the human-computer interaction module. At this time, the total output of the production line cannot meet the production demand, that is, the production task cannot be met;

[0043] The production optimization module receives the optimization instruction sent by the comparison module, generates a production plan, obtains the optimal production plan based on the genetic algorithm, and sends a control instruction to the adjustment control module;

[0044] The adjustment control module is used to receive data from the real-time monitoring module and control instructions from the production optimization module to adjust the production equipment accordingly;

[0045] The human-computer interaction module is used to perform human-computer interaction display on the data.

[0046] In this embodiment, it should be specifically explained that the specific method of the real-time monitoring module is:

[0047] If the temperature of the production equipment exceeds the preset temperature threshold, a temperature warning notification is sent to the human-computer interaction module, and temperature adjustment data is sent to the adjustment control module at the same time; at this time, the production equipment will malfunction due to excessive temperature and thus cannot operate normally, thereby affecting production, so it is necessary to timely control the temperature of the production equipment; the preset temperature threshold is the maximum operating temperature value of the production equipment; the temperature adjustment data is the temperature adjustment value, that is, the value of the production equipment temperature exceeding the preset temperature threshold;

[0048] If the pressure of the production equipment exceeds the preset pressure threshold, a pressure warning notification is sent to the human-computer interaction module, and temperature adjustment data is sent to the adjustment control module at the same time; at this time, the production equipment will malfunction due to excessive pressure and will not be able to operate normally, thereby affecting production, so it is necessary to regulate the pressure of the production equipment in time; the preset pressure threshold is the maximum operating pressure value of the production equipment; the pressure adjustment data is the pressure adjustment value, that is, the value of the production equipment pressure exceeding the preset pressure threshold.

[0049] In this embodiment, it should be specifically stated that the neural network model is an LSTM neural network model, and the construction method includes:

[0050] Initialize the basic architecture of the LSTM neural network model, which includes an input layer, two LSTM units, a Dropout layer, a fully connected layer, a learnable parameter activation function, a comprehensive loss function, an optimizer, a regularizer, and an output layer; the output layer includes n nodes;

[0051] Each LSTM unit consists of a memory unit and three control gates; the three control gates are input gate, forget gate and output gate; a fully connected layer is added between LSTM units;

[0052] The Dropout layer randomly discards a certain proportion of nodes from the LSTM neural network model temporarily, which can effectively prevent overfitting;

[0053] The learnable parameter activation function is: Where s is the input data; p1 is a learnable slope parameter that controls the slope of the linear transformation of s. The value of p1 affects the gradient of the Sigmoid function in the input space, thereby affecting the response speed of the Sigmoid function to the input; learning p1 can make the model adapt to different ranges of activation values ​​in the data; p2 is a learnable offset parameter that controls the offset of the Sigmoid function, enabling the model to adapt to different baselines or offsets in the input data; learning p2 can enable the model to better model the overall pattern of the data;

[0054] The specific values ​​of p1 and p2 are obtained through model training; in deep learning, training is performed through the back-propagation algorithm and optimizer, which adjusts the parameters in the network so that the model achieves better performance on the training data;

[0055] Uncertainty modeling is introduced into the loss function to obtain the comprehensive loss function Loss:

[0056]

[0057] ; Where q is the quantile, which indicates the quantile level in the loss calculation; q takes a value between (0, 1), which determines whether the loss is calculated for the lower quantile or the upper quantile; y true is the real target value, indicating the actual production line output; is the lower quantile of the predicted distribution; is the upper quantile of the predicted distribution; | | + is the positive function; it is used to ensure that only when the difference between the predicted value and the true value is positive will it contribute to the loss;

[0058] The optimizer adopts an adaptive optimizer that adaptively adjusts the learning rate;

[0059] The optimization formula of the adaptive optimizer is: in, is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ+1; is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ; η l is the learning rate of the lth layer of the LSTM neural network model, and different learning rates are selected according to different layers; v λ is the exponential moving average of the squared gradients for training iterations λ; θ is a small constant to prevent the denominator from being zero; m λ，l is the parameter gradient for the training iteration number λ;

[0060] like but in is the value of the comprehensive loss function when the number of training iterations is λ; is the value of the comprehensive loss function when the number of training iterations is λ-1; A is the reduction parameter;

[0061] The regularizer uses L2 norm regularization to control the complexity of the LSTM neural network model and prevent overfitting;

[0062] The n output layers are used to output the production output of n production lines;

[0063] If there are n production lines in total, the production output of the n production lines over a period of time in history is collected to construct training data; the training data is divided into a training set and a validation set; the specific division ratio can be set by those skilled in the art, and this embodiment selects 80% training set and 20% validation set;

[0064] The training set is output to the LSTM neural network model, and the n production lines correspond to the n nodes of the output layer. During the training process, when the function value of the comprehensive loss function does not decrease during the process of P consecutive iterations, the training is completed, and the trained LSTM neural network model is obtained.

[0065] The collected output of n production lines is input into the trained LSTM neural network model to obtain the production output of n production lines in the future t time period.

[0066] In this embodiment, it should be specifically explained that the specific manner in which the comparison module compares the predicted production output with the production demand is:

[0067] Calculate total production: Among them, Z is the total output, C ij is the output of the i-th production line at the j-th moment, j = 1, 2, 3...t, i = 1, 2, 3...n;

[0068] If the total output at time t reaches the production demand at time t, an optimization instruction is sent to the production optimization module; at this time, the original production plan of the production line can meet the production demand, so the production plan can be optimized and adjusted to achieve optimal production.

[0069] In this embodiment, it should be specifically explained that the objective function of the production optimization module to generate the production plan is: Among them, Z′ is the production demand, that is, the production output to be achieved; C ij is the output of the i-th production line at the j-th moment, j = 1, 2, 3...t, i = 1, 2, 3...n; all C ij Both can be used as production solutions;

[0070] The genetic algorithm to obtain the optimal production plan includes the following steps:

[0071] Step S01: Encode the production plan, obtain chromosomes, and construct the initial population;

[0072] Step S02: Determine the fitness function;

[0073] Step S03: Perform natural selection on chromosomes in the population;

[0074] Step S04: performing crossover recombination on chromosomes in the population;

[0075] Step S05: mutating the chromosomes in the population;

[0076] Step S06: obtain a new population, the preset population generation number is L, the fitness threshold is Q, L is an integer greater than 0, and Q is a real number greater than 0; loop step S03-step S05 until the generation number corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is taken as the optimal production plan; illustratively, if the preset population generation number is 1, the chromosomes in the initial population are subjected to natural selection, crossover recombination and mutation to obtain a new population, and the generation number corresponding to the new population is 1 at this time, so the loop ends;

[0077] The fitness threshold Q is preset by those skilled in the art according to the algorithm accuracy. The population algebra L is obtained by those skilled in the art using the genetic algorithm multiple times under multiple different production plans to obtain the optimal production plan. In each process of using the genetic algorithm, when the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the cycle ends and the algebra corresponding to the new population is obtained; the largest algebra among the multiple algebras is taken as the population algebra L;

[0078] In step S01, the production plan is encoded as U, where U is a chromosome; N chromosomes are randomly generated to form an initial population Z0 = {U1, U2, ..., U N};

[0079] In step S02, the fitness function is expressed as: r =-COS r ;

[0080] In the formula, f r is the fitness corresponding to the rth chromosome, COS r is the production cost corresponding to the rth chromosome, r∈[1,N]; the production cost corresponding to the rth chromosome is the cost incurred by the operation of all production lines in the rth production plan, including but not limited to electricity cost and water resource cost. The specific cost content can be set according to the actual production situation of the product;

[0081] In step S03, natural selection is performed by combining the elite method and the rotation method; the elite method generates k1 offspring chromosomes, and for a population with a capacity of N, the fitness corresponding to the N chromosomes is arranged from large to small, and the k1 chromosomes at the front each generate a offspring chromosome; the rotation method generates k2 offspring chromosomes, that is, N chromosomes generate k2 offspring chromosomes according to the corresponding rotation probability; k1+k2=N;

[0082] The expression of rotation probability is:

[0083] Where P r is the rotation probability corresponding to the rth chromosome;

[0084] In the step S04, G chromosomes are randomly selected from the population for crossover recombination to obtain G new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the G new chromosomes is calculated, the fitness of the G new chromosomes and the fitness of the G chromosomes are sorted from large to small and a sorting table is generated, and the G chromosomes in the sorting table are replaced with the G chromosomes in the population for crossover recombination in positive order; in this embodiment, G=0.7N is selected;

[0085] In step S05, the mutation probability is preset to T, and the N chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes. In this embodiment, T=0.02 is selected;

[0086] In this embodiment, it should be specifically explained that the adjustment control module is used to receive data from the real-time monitoring module and adjust the temperature and pressure of the production equipment based on the temperature adjustment value and the pressure adjustment value;

[0087] Based on the control instructions of the production optimization module, the production equipment is regulated and controlled. The production equipment of each production line is regulated according to the corresponding production output in the production plan. On the basis of meeting the production demand, the production cost is minimized to improve industrial production efficiency and profits.

[0088] In the present embodiment, it should be specifically explained that the difference between the present embodiment and the prior art lies mainly in that the present embodiment is provided with a production prediction module and a production optimization module. Based on the constructed neural network model, the output of the production line is predicted to obtain the predicted production output. If the production demand can be met, the production plan is further optimized, and the optimal production plan is obtained based on the genetic algorithm, so as to minimize the production cost, improve the intelligence of industrial production, and reduce the industrial production cost. At the same time, a real-time monitoring module is provided to monitor the key parameters such as the temperature and pressure of the production equipment in real time to prevent the production equipment from malfunctioning and causing the production line to be unable to operate, thereby reducing the industrial production efficiency.

[0089] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0090] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An industrial production control system based on industrial Internet of Things, characterized by: It includes data acquisition module, data preprocessing module, real-time monitoring module, production prediction module, comparison module, production optimization module, adjustment control module and human-computer interaction module; The data acquisition module is used to collect production equipment data and transmit it to the data preprocessing module; The data preprocessing module is used to preprocess the data collected by the data acquisition module to obtain production equipment data that can be directly used, and transmit it to the real-time monitoring module and the production prediction module; The real-time monitoring module is used to receive data from the data preprocessing module, monitor the temperature and pressure of the production equipment in real time, and send an early warning notification to the human-computer interaction module if the set threshold is exceeded; The production prediction module predicts the output of the production line based on the constructed neural network model to obtain the predicted production output; The comparison module compares the predicted production output of the production prediction module with the production demand. If the production demand is reached, an optimization instruction is sent to the production optimization module; otherwise, an early warning prompt is sent to the human-computer interaction module. The production optimization module receives the optimization instruction sent by the comparison module, generates a production plan, obtains the optimal production plan based on the genetic algorithm, and sends a control instruction to the adjustment control module; The adjustment control module is used to receive data from the real-time monitoring module and control instructions from the production optimization module to adjust the production equipment accordingly; The human-computer interaction module is used to perform human-computer interaction display on the data.

2. The industrial production control system based on industrial Internet of Things according to claim 1 is characterized in that: The neural network model is an LSTM neural network model, and the construction method includes: Initialize the basic architecture of the LSTM neural network model, which includes an input layer, two LSTM units, a Dropout layer, a fully connected layer, a learnable parameter activation function, a comprehensive loss function, an optimizer, a regularizer, and an output layer; the output layer includes n nodes; Each LSTM unit consists of a memory unit and three control gates; the three control gates are input gate, forget gate and output gate; a fully connected layer is added between LSTM units; The n output layers are used to output the production output of the n production lines.

3. The industrial production control system based on industrial Internet of Things according to claim 2 is characterized in that: The learnable parameter activation function is: Where s is the input data; p1 is a learnable slope parameter that controls the slope of the linear transformation of s; p2 is a learnable offset parameter that controls the offset of the Sigmoid function; Uncertainty modeling is introduced into the loss function to obtain the comprehensive loss function Loss: ； Among them, q is the quantile, which represents the quantile level in the loss calculation; q takes a value between (0, 1); y true is the real target value, indicating the actual production line output; is the lower quantile of the predicted distribution; is the upper quantile of the predicted distribution; | |+ is the positive function; The optimizer adopts an adaptive optimizer that adaptively adjusts the learning rate.

4. The industrial production control system based on industrial Internet of Things according to claim 3 is characterized in that: The optimization formula of the adaptive optimizer is: in, is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ+1; is the parameter of the lth layer of the LSTM neural network model when the number of training iterations is λ; η l is the learning rate of the lth layer of the LSTM neural network model, and different learning rates are selected according to different layers; v λ is the exponential moving average of the squared gradients for training iterations λ; θ is a small constant to prevent the denominator from being zero; m λ，l is the parameter gradient for the number of training iterations λ.

5. The industrial production control system based on industrial Internet of Things according to claim 3 is characterized in that: In the loss function, if but in is the value of the comprehensive loss function when the number of training iterations is λ; is the value of the comprehensive loss function when the number of training iterations is λ-1; A is the reduction parameter.

6. The industrial production control system based on industrial Internet of Things according to claim 1 is characterized in that: The objective function of the production optimization module to generate the production plan is: Among them, Z′ is the production demand, that is, the production output to be achieved; C ij is the output of the i-th production line at the j-th moment, j = 1, 2, 3...t, i = 1, 2, 3...n.

7. The industrial production control system based on industrial Internet of Things according to claim 1 is characterized in that: The genetic algorithm to obtain the optimal production plan includes the following steps: Step S01: Encode the production plan, obtain chromosomes, and construct the initial population; Step S02: Determine the fitness function; Step S03: Perform natural selection on chromosomes in the population; Step S04: performing crossover recombination on chromosomes in the population; Step S05: mutating the chromosomes in the population; Step S06: Obtain a new population, preset the population generation to be L, the fitness threshold to be Q, L is an integer greater than 0, and Q is a real number greater than 0; loop steps S03 to S05 until the generation corresponding to the new population is L or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold Q, the loop ends, and the production plan corresponding to the chromosome with the maximum fitness in the new population is taken as the optimal production plan.

Images