CN115081703A

CN115081703A - Gas quantity prediction method, device, equipment and storage medium

Info

Publication number: CN115081703A
Application number: CN202210679786.9A
Authority: CN
Inventors: 宋骄; 许号; 吴铁成; 程堂灿
Original assignee: Suzhou Luster Vision Intelligent Device Co Ltd
Current assignee: Suzhou Luster Vision Intelligent Device Co Ltd
Priority date: 2022-06-15
Filing date: 2022-06-15
Publication date: 2022-09-20

Abstract

The invention discloses a gas quantity prediction method, a gas quantity prediction device, gas quantity prediction equipment and a storage medium. The method comprises the following steps: acquiring attribute information and offset printing speed information of a product to be printed; inputting the attribute information and the offset printing speed information of the product to be printed into a gas quantity prediction model to obtain a gas quantity prediction result, and supplying gas according to the gas quantity prediction result; the gas quantity prediction model is obtained by training based on pre-acquired sample data, and the sample data is determined based on a detection result of a product image acquired in an offset printing process. According to the technical scheme, the air quantity required in the offset printing online detection process is accurately predicted through the air quantity prediction model, the problem that the single air quantity is difficult to meet the air quantity requirements of different products in the offset printing online detection process can be solved, and the purposes of saving energy and reducing production cost are achieved while the air quantity supply adaptability is improved.

Description

Gas quantity prediction method, device, equipment and storage medium

Technical Field

The invention relates to the technical field of data processing, in particular to a gas quantity prediction method, a gas quantity prediction device, gas quantity prediction equipment and a storage medium.

Background

With the continuous progress of printing inspection technology, quality inspection has penetrated from a single quality inspection stage to each process of printing. The introduction of offset printing on-line detection is beneficial to improving the quality of printed matters while saving a large amount of paper for users.

At present, in an offset printing online detection scheme, a product to be printed runs at a high speed along with a large roller, and in order to enable an image collected by a camera to be clearer, a larger air flow is required to blow the foremost end of the product. Printing plants typically use a single, larger volume of gas to meet the volume requirements of the various products to be printed.

However, due to different attributes of the product to be printed, such as gram weight, breadth and the like, a single gas amount cannot be adapted to meet the gas amount requirements of different products, the power consumption is greatly increased, high electricity cost is brought, and the production cost is further increased.

Disclosure of Invention

The invention provides a gas quantity prediction method, a gas quantity prediction device, gas quantity prediction equipment and a storage medium, which are used for solving the problem that a single gas quantity is difficult to meet the gas quantity requirements of different products in the offset printing online detection process, and can achieve the purposes of saving energy and reducing production cost while improving the gas quantity supply adaptability.

According to an aspect of the present invention, there is provided a gas amount prediction method, including:

acquiring attribute information and offset printing speed information of a product to be printed;

inputting the attribute information and the offset printing speed information of the product to be printed into a gas quantity prediction model to obtain a gas quantity prediction result, and supplying gas according to the gas quantity prediction result; the gas quantity prediction model is obtained by training based on pre-acquired sample data, and the sample data is determined based on a detection result of a product image acquired in an offset printing process.

According to another aspect of the present invention, there is provided a gas amount prediction apparatus, including:

the information acquisition module is used for acquiring attribute information and offset printing speed information of a product to be printed;

the air volume prediction result determining module is used for inputting the attribute information and the offset printing speed information of the product to be printed into an air volume prediction model to obtain an air volume prediction result so as to supply air according to the air volume prediction result; the gas quantity prediction model is obtained by training based on pre-acquired sample data, and the sample data is determined based on a detection result of a product image acquired in an offset printing process.

According to another aspect of the present invention, there is provided an electronic apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to enable the at least one processor to perform the gas quantity prediction method according to any of the embodiments of the present invention.

According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to implement the gas quantity prediction method according to any one of the embodiments of the present invention when executed.

According to the technical scheme of the embodiment of the invention, the air quantity required in the offset printing online detection process is accurately predicted through the air quantity prediction model, the problem that the air quantity requirements of different products are difficult to meet by a single air quantity in the offset printing online detection process can be solved, and the purposes of saving energy and reducing production cost are achieved while the air quantity supply adaptability is improved.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present invention, nor do they necessarily limit the scope of the invention. Other features of the present invention will become apparent from the following description.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart of a gas amount prediction method according to an embodiment of the present invention;

FIG. 2 is a flowchart of a method for training a gas quantity prediction model according to a second embodiment of the present invention;

FIG. 3A is a flow chart of a gas quantity prediction method provided according to a first embodiment of the present invention;

FIG. 3B is a schematic diagram illustrating a gas quantity prediction model provided in a first scenario in which the present invention is specifically applied;

FIG. 3C is a schematic diagram illustrating a gene encoding method for a regulatory parameter according to a first embodiment of the present invention;

fig. 4 is a schematic structural diagram of a gas quantity prediction apparatus according to a third embodiment of the present invention;

fig. 5 is a schematic structural diagram of an electronic device implementing the gas amount prediction method according to the embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. According to the technical scheme, the data acquisition, storage, use, processing and the like meet relevant regulations of national laws and regulations.

Example one

Fig. 1 is a flowchart of a gas amount prediction method according to an embodiment of the present invention, which is applicable to a gas amount prediction scenario, and is particularly applicable to a gas amount prediction situation in an offset printing online detection scenario. The method may be performed by a gas amount prediction apparatus, which may be implemented in hardware and/or software, and may be configured in an electronic device. As shown in fig. 1, the method includes:

and S110, acquiring attribute information and offset printing speed information of a product to be printed.

The scheme can be executed by an offset printing online detection system, and the offset printing online detection system can be configured on offset printing equipment to monitor the quality of printed products, the running state of the offset printing equipment and the like in real time. The offset printing online detection system can read attribute information of a product to be printed, such as the grammage of paper, the length of paper, the width of paper, and the like, which are input in advance or stored in the system. The offset printing online detection system can be communicated with offset printing equipment to acquire attribute information of a product to be printed in real time. The offset printing online detection system can continuously acquire offset printing speed information in the running process of offset printing equipment through speed acquisition equipment such as an encoder.

S120, inputting the attribute information and the offset printing speed information of the product to be printed into a gas quantity prediction model to obtain a gas quantity prediction result, and feeding gas according to the gas quantity prediction result; the gas quantity prediction model is obtained by training based on pre-acquired sample data, and the sample data is determined based on a detection result of a product image acquired in an offset printing process.

The offset printing online detection system can input attribute information and offset printing speed information of a product to be printed into the air volume prediction model as input data of the air volume prediction model. It is easy to understand that the offset air consumption has a correlation with the properties of the product to be printed, and is also influenced by the running speed of the offset printing equipment. Therefore, a mapping relation exists among the attribute information of the product to be printed, the offset printing speed information and the offset printing air consumption, and an air quantity prediction model can be used for describing the mapping relation. The air quantity prediction model can output air quantity prediction results according to the attribute information of the product to be printed and the offset printing speed information. The offset printing online detection system can supply air in the offset printing process through the air supply equipment according to the air quantity prediction result so as to ensure that the offset printing online detection obtains a clear product image.

The gas amount prediction model may be obtained by training based on pre-acquired sample data, and may be a deep learning model such as a neural network. In order to better meet the requirements of practical application scenes, the offset printing online detection system can extract the attribute information of the printed product, the offset printing speed information and the offset printing air consumption in the historical record as sample data. And taking the attribute information and the offset printing speed information of the printed product as independent variables, taking the offset printing air consumption as dependent variables, and constructing a mapping relation between the independent variables and the dependent variables so as to obtain an air quantity prediction result according to the mapping relation after inputting the attribute information and the offset printing speed information of the product to be printed. It should be noted that the selection of the sample data has a certain standard, specifically, the offset printing online detection system may perform air blowing quality detection based on a product image acquired in an offset printing process, and the sample data may be selected based on an air blowing quality detection result.

According to the technical scheme, the air quantity required in the offset printing online detection process is accurately predicted through the air quantity prediction model, the problem that the single air quantity is difficult to meet the air quantity requirements of different products in the offset printing online detection process can be solved, and the purposes of saving energy and reducing production cost are achieved while the air quantity supply adaptability is improved.

Example two

Fig. 2 is a flowchart of a training method of a gas quantity prediction model according to a second embodiment of the present invention, which is detailed based on the second embodiment. As shown in fig. 2, the training method of the gas amount prediction model may include:

s210, collecting a product image in the offset printing process according to a preset sampling frequency, and synchronously acquiring gas flow information.

In the scheme, the gas quantity prediction model is constructed based on a Radial Basis Function (RBF) neural network. The offset online inspection system may include an image capture device, such as a camera. In the running process of offset printing equipment, an offset printing online detection system can acquire a product image in the offset printing process according to a certain sampling frequency. The offset printing online detection system can also comprise a position information acquisition device, such as a photoelectric sensor, for acquiring the position information of the product in the offset printing process so as to acquire the product image within a reasonable shooting range. The offset in-line inspection system may also include a gas flow collection device, such as a flow meter. When the product image is collected, the offset printing online detection system can synchronously acquire gas flow information through the flow collection equipment.

S220, determining at least one group of sample data according to the detection result of the product image; wherein the sample data includes product attribute information associated with the product image, pre-acquired offset printing speed information, and the gas flow information.

The offset printing online detection system can use product attribute information associated with a product image, offset printing speed information of offset printing equipment operation and gas flow information synchronously acquired with the product image as a group of sample data, and a sample database is constructed by acquiring multiple groups of sample data so as to train a gas amount prediction model by using the sample data in the sample database.

It should be noted that, when the offset printing online detection system trains the air volume prediction model, a large amount of sample data, for example, hundreds of sets of sample data, may be selected for training to increase the model adaptability. When the air quantity prediction model has structural advantages, such as a deep learning structure based on single-sample learning or few-sample learning, the offset printing online detection system can also select a small amount of sample data, such as several groups or dozens of groups of sample data, even one group of sample data, for training.

During the sample collection process, the air volume provided by the air supply device during the offset printing process may not meet the requirement of the offset printing online detection system during the product image collection. In order to ensure the reliability of sample data, a gas quantity prediction model with good training effect is obtained. The offset printing online detection system can select the blowing air amount information which enables the product to be well flattened as sample data according to the detection result of the product image. Specifically, the offset printing online detection system can perform target detection on a product image, the detected target can be whether areas such as product tilting and rolling exist in the product image, and the offset printing online detection system can also judge whether the air flow is appropriate by detecting whether changes in appearance exist in two adjacent product images, for example, whether the paper length and the paper width in the two images of the same product are consistent.

In this scheme, optionally, the product image is a three-dimensional image;

correspondingly, the determining at least one set of sample data according to the detection result of the product image includes:

detecting whether a product has a flattening abnormal area or not according to the three-dimensional image;

and if the flattening abnormal area does not exist, taking product attribute information associated with the three-dimensional image, offset printing speed information acquired in advance and gas flow information acquired synchronously with the three-dimensional image as a group of sample data.

The offset printing online detection system can shoot a product in an offset printing process at multiple angles to obtain a three-dimensional image of the product. According to the three-dimensional image, the offset printing online detection system can more easily judge whether the product has flattening abnormal areas such as pits, bulges and the like in the offset printing process. If the three-dimensional image detects that the product is well flattened and no abnormal area exists, the offset printing online detection system can take product attribute information associated with the three-dimensional image, offset printing speed information acquired in advance and gas flow information acquired synchronously with the three-dimensional image as a group of sample data for training a gas amount prediction model. The product attribute information and the offset printing speed information can be used as input data of the gas quantity prediction model, and the gas flow information is used as label data of the gas quantity prediction model.

According to the scheme, the abnormal area of the product flattening can be quickly positioned through the three-dimensional image, the reliability of sample data selection is improved, and a good model training effect is favorably realized.

And S230, training the radial basis function neural network by using the sample data to obtain a gas quantity prediction model.

The offset online inspection system may input product attribute information and offset speed information to the radial basis function neural network. The structure of the radial basis function neural network is similar to that of the multilayer forward network, and the radial basis function neural network can be a three-layer forward network. The input layer is composed of signal source nodes, for example, product attribute information and offset printing speed information can be used as two signal source nodes of the input layer. The second layer is a hidden layer, the number of hidden units can be determined according to specific application requirements, and a transform function RBF of the hidden units is a nonnegative nonlinear function which is radially symmetrical to a central point and is attenuated. The third layer is an output layer that can respond to the effects of the input data. The transformation from the input space to the hidden layer space is non-linear, while the output layer spatial transformation from the hidden layer space is linear. The radial basis function neural network can extract features in input data, output gas quantity prediction features, and optimize internal weight parameters of the radial basis function neural network by not calculating a difference between gas flow information and the gas quantity prediction features so as to obtain a gas quantity prediction model.

On the basis of the foregoing scheme, optionally, training the radial basis function neural network by using the sample data to obtain a gas quantity prediction model, including:

inputting the product attribute information and the offset printing speed information into a radial basis function neural network as model input data to obtain a gas quantity prediction characteristic;

and optimizing the radial basis function neural network according to the gas quantity prediction characteristics and the gas flow information until preset conditions are met to obtain a gas quantity prediction model.

It can be understood that the gas amount prediction feature may be output data of the radial basis function neural network in a training process, and the offset printing online detection system may encode the gas flow information into a data format of the same type as the gas amount prediction feature. The radial basis function neural network can take the gas flow information as the final purpose of training, calculate the difference data between the gas flow prediction characteristic and the gas flow information through an activation function, feed back the difference data to a hidden layer of the radial basis function neural network when the difference data does not meet the preset condition so as to continuously optimize the weight parameter, and output the gas flow prediction model until the difference data meets the preset condition. The activation function may be a gaussian function, an inverse sigmoid function, a quasi-quadratic function, or the like. The preset condition can be that the training iteration number reaches the preset iteration number, and whether the loss value, the accuracy rate and the like in the radial basis function neural network training process meet preset evaluation indexes or not.

According to the scheme, the gas quantity prediction model is obtained by training the radial basis function neural network, so that the reliable prediction of the gas quantity is realized, and the robustness of the gas quantity prediction model is improved.

In one possible solution, optionally, the gas amount prediction model includes a tuning parameter;

after obtaining the gas quantity prediction model, the method further comprises:

optimizing the adjusting parameters through a genetic algorithm until an optimization termination condition is met, and outputting an optimization result;

and updating the adjusting parameters of the gas quantity prediction model according to the optimization result.

The offset printing online detection system can optimize the adjusting parameters in the air quantity prediction model by utilizing a genetic algorithm so as to obtain a more accurate air quantity prediction result. The adjusting parameters may include a center vector, a base width vector, a weight vector, and the like in the radial basis function neural network. Through a genetic algorithm, the offset printing online detection system can optimize one type of adjusting parameters independently or simultaneously. The optimization termination condition may be that the iteration number of the genetic algorithm reaches a preset iteration number.

The scheme can further optimize the adjusting parameters of the air quantity prediction model, and is favorable for realizing high-precision prediction of offset air quantity.

On the basis of the above scheme, optionally, the optimizing the adjustment parameter by using a genetic algorithm until an optimization termination condition is met, and outputting an optimization result, including:

carrying out gene coding on the regulating parameters to generate an input population;

carrying out selection operation, cross operation and variation operation on the input population to obtain an output population;

calculating the fitness of individuals in the output population, and determining the input population of the next iteration according to the fitness calculation result;

and if the iteration times reach the preset iteration times, terminating the iteration, and determining the optimization result of the adjusting parameter according to the finally obtained output population.

Specifically, the offset printing online detection system may genetically encode the regulatory parameters to be optimized to obtain an initial population, i.e., an input population for a first iteration. And selecting, crossing and mutating the initial population to realize population amplification and obtain an output population. And selecting the individuals from the output population to enter the input population of the next iteration by calculating the fitness of the individuals in the output population and comparing the result of the fitness calculation with the preset selection condition. And terminating the iteration until the iteration times reach the preset iteration times, and outputting an optimization result of the adjusting parameters.

According to the scheme, the optimization result of the adjusting parameters is output by simulating the evolution process of the population, so that the adjusting parameters which are more fit with the application scene can be obtained, and the prediction accuracy of the gas quantity prediction model is improved.

In a preferred aspect, the adjustment parameters include a center vector and a base width vector;

the genetic encoding of the regulatory parameters to generate an input population comprises:

the individuals are genetically encoded in such a way that elements in the center vector alternate with elements in the base width vector to generate an input population.

In the scheme, the offset printing online detection system can simultaneously optimize two types of adjusting parameters of a center vector and a base width vector through a genetic algorithm. In the gene coding process, each element in the central vector and each element in the base width vector are coded in an alternating mode, so that the frequency and the probability of exchanging the central data and the width data can be improved, the diversity of the population can be increased, and the overall searching capability of the genetic algorithm can be improved.

Specific application scenario 1

Fig. 3A is a flowchart of a gas amount prediction method according to a first embodiment of the present invention, which is a specific embodiment based on the foregoing embodiments. As shown in fig. 3A, the method includes:

step 1: sample collection, namely inputting the specification of a product to be printed into a data acquisition system; when the offset press runs, the encoder acquires the offset printing running speed in real time and inputs the speed value into the data acquisition system; when the offset press is normally produced, the flow meter collects the gas consumption in real time and transmits the gas consumption to the data acquisition system through RS485 communication. And the data acquisition system collects the specification of a product to be printed, the offset printing running speed and the air consumption into a sample library.

Step 2: data verification, namely screening the data of the sample library and deleting abnormal data; and when the offset press is normally produced, the image acquisition detection is carried out on the product under the gas consumption counted by the flow meter. If the blowing flattening effect is normal, the gas flow counted at the moment is true and is reserved as sample data, and if the blowing flattening effect is abnormal, the gas flow is counted as false and cannot be used as the sample data.

And step 3: and eliminating invalid sample data in the sample library, dividing the invalid sample data into a training library and a testing library, and training the RBF network by using the sample library to obtain a preliminary model.

And 4, step 4: and (3) optimizing parameters of the RBF network by using a Genetic Algorithm (GA), so as to obtain an optimized GA-RBF network model.

And 5: and predicting the air consumption of the offset press by using the optimized GA-RBF network model, comparing the air consumption with the output air consumption (namely the actual air consumption) of the test library, and if the error is smaller than a preset error threshold, conforming to the optimization termination condition.

The specific application mode of the GA-RBF network model is as follows:

1. RBF neural network

Fig. 3B is a schematic structural diagram of a gas amount prediction model provided according to a first specific application scenario of the present invention. In the scheme, the RBF neural network selects a Gaussian function as a hidden layer kernel function.

The output of the hidden layer is a nonlinear activation function h _i The method comprises the following steps:

where X represents the input sample vector, C _i Represents a center vector; b _i Is the width of the basis function; and m is the number of nodes of the hidden layer.

The weight value between a hidden layer and an output layer in the RBF network and the output threshold value are continuously adjusted under the supervision of a supervised learning rule. The key of the RBF network is the determination of a center vector, a base width vector and a weight vector.

Because the input of the RBF network is a four-dimensional array, and the output is a one-dimensional numerical value, the input and the output have a nonlinear relationship. The nonlinear approximation function of RBF networks is well suited to solve the theoretically existing nonlinear relationships.

2. Genetic algorithm optimized RBF network

Determining a data center vector c when designing an RBF network _k Base width vector b _k And a weight vector w _k The specific values of the three parameters are the key for improving the speed and the precision of the RBF network in solving the problems. The genetic algorithm optimizes network parameters and comprises the following aspects:

(1) encoding and population initialization

FIG. 3C is a schematic diagram of a gene encoding method for a regulatory parameter according to a first embodiment of the present invention. The scheme mainly embodies the following two aspects of the optimization of the traditional RBF by selecting a real number coding mode:

the first point is as follows: the traditional RBF network design is completed step by step, the first step is k-means clustering method to select the data center and the basic width constant of the network, the second step is least square method to solve the edge weight, and then the network output is obtained. Thus, the solutions are calculated step by step, the optimal solution is not necessarily synchronous, and the superiority of the algorithm is influenced. The genetic algorithm puts the codes of the three types of parameters into a chromosome coding string, thereby realizing the synchronism of the optimal solution of the parameters.

And a second point: FIG. 3C shows the interplay between data centers and basis function widths in an RBF neural network. C in chromosomes to get closer to the radial basis function itself _k And b _k Alternate, with w rearranged at the end of the chromosome string _k . The individuals are encoded using the above arrangement. The method is beneficial to improving the exchange frequency and probability of the center and the width, is vital to the diversity of the population, and can improve the overall searching capability of the network.

(2) Parameter optimization update

Let the output of the RBF network be:

where X represents the input sample vector, C _i Representing a central vector, w _i Representing the output weights of the hidden layer nodes.

The energy function of the RBF network is:

wherein, X ^p Representing a vector of input samples; n denotes the number of training samples, d _p Indicating an output expectation.

C _i And b _i The update of (1) is:

where t denotes the number of iterations, λ is the learning efficiency of the basis function center value, and β is the learning efficiency of the basis function width value.

EXAMPLE III

Fig. 4 is a schematic structural diagram of a gas amount prediction device according to a third embodiment of the present invention. As shown in fig. 4, the apparatus includes:

the information acquisition module 410 is used for acquiring attribute information and offset printing speed information of a product to be printed;

the air volume prediction result determining module 420 is configured to input the attribute information of the product to be printed and the offset printing speed information into an air volume prediction model to obtain an air volume prediction result, and to supply air according to the air volume prediction result; the gas quantity prediction model is obtained by training based on pre-acquired sample data, and the sample data is determined based on a detection result of a product image acquired in an offset printing process.

In this embodiment, optionally, the apparatus further includes:

the image and flow acquisition module is used for acquiring a product image in the offset printing process according to a preset sampling frequency and synchronously acquiring gas flow information;

the sample data determining module is used for determining at least one group of sample data according to the detection result of the product image; wherein the sample data comprises product attribute information associated with the product image, pre-acquired offset printing speed information, and the gas flow information;

and the gas quantity prediction model generation module is used for training the radial basis function neural network by using the sample data to obtain a gas quantity prediction model.

In one possible solution, optionally, the product image is a three-dimensional image;

correspondingly, the sample data determining module includes:

the anomaly detection unit is used for detecting whether the product has a flattening anomaly region according to the three-dimensional image;

and the sample data determining unit is used for taking the product attribute information associated with the three-dimensional image, the offset printing speed information acquired in advance and the gas flow information acquired synchronously with the three-dimensional image as a set of sample data if the flattening abnormal area does not exist.

On the basis of the foregoing scheme, optionally, the gas amount prediction model generation module includes:

the air volume prediction characteristic generation unit is used for inputting the product attribute information and the offset printing speed information into a radial basis function neural network as model input data to obtain an air volume prediction characteristic;

and the gas quantity prediction model generation unit is used for optimizing the radial basis function neural network according to the gas quantity prediction characteristics and the gas flow information until preset conditions are met so as to obtain a gas quantity prediction model.

In another possible implementation, the gas quantity prediction model includes tuning parameters;

the device further comprises:

the optimization result output module is used for optimizing the adjusting parameters through a genetic algorithm until the optimization termination condition is met and outputting an optimization result;

and the adjusting parameter updating module is used for updating the adjusting parameters of the gas quantity prediction model according to the optimization result.

On the basis of the above scheme, optionally, the optimization result output module includes:

the input population generating unit is used for carrying out gene coding on the regulating parameters to generate an input population;

the output population generating unit is used for carrying out selection operation, cross operation and variation operation on the input population to obtain an output population;

the fitness calculation unit is used for calculating the fitness of individuals in the output population and determining the input population of the next iteration according to the fitness calculation result;

and the optimization result output unit is used for terminating the iteration if the iteration times reach the preset iteration times and determining the optimization result of the adjusting parameter according to the finally obtained output population.

the input population generating unit is specifically configured to perform gene coding on the individuals in such a manner that each element in the center vector and each element in the base width vector alternate, so as to generate the input population.

The gas quantity prediction device provided by the embodiment of the invention can execute the gas quantity prediction method provided by any embodiment of the invention, and has the corresponding functional module and beneficial effect of the execution method.

Example four

FIG. 5 illustrates a schematic diagram of an electronic device 510 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 5, the electronic device 510 includes at least one processor 511, and a memory communicatively connected to the at least one processor 511, such as a Read Only Memory (ROM)512, a Random Access Memory (RAM)513, and the like, wherein the memory stores a computer program executable by the at least one processor, and the processor 511 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM)512 or the computer program loaded from a storage unit 518 into the Random Access Memory (RAM) 513. In the RAM 513, various programs and data necessary for the operation of the electronic device 510 can also be stored. The processor 511, the ROM 512, and the RAM 513 are connected to each other by a bus 514. An input/output (I/O) interface 515 is also connected to bus 514.

Various components in the electronic device 510 are connected to the I/O interface 515, including: an input unit 516 such as a keyboard, a mouse, and the like; an output unit 517 such as various types of displays, speakers, and the like; a storage unit 518, such as a magnetic disk, optical disk, or the like; and a communication unit 519 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 519 allows the electronic device 510 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks.

Processor 511 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 511 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, or the like. The processor 511 performs various methods and processes described above, such as a gas quantity prediction method.

In some embodiments, the gas quantity prediction method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 518. In some embodiments, some or all of the computer program may be loaded and/or installed onto the electronic device 510 via the ROM 512 and/or the communication unit 519. When loaded into RAM 513 and executed by processor 511, may perform one or more of the steps of the gas quantity prediction method described above. Alternatively, in other embodiments, the processor 511 may be configured to perform the gas amount prediction method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. A computer program can execute entirely on a machine, partly on a machine, as a stand-alone software package partly on a machine and partly on a remote machine or entirely on a remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired results of the technical solution of the present invention can be achieved.

The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A gas amount prediction method, characterized in that the method comprises:

2. The method of claim 1, wherein the training process of the gas quantity prediction model comprises:

collecting a product image in the offset printing process according to a preset sampling frequency, and synchronously acquiring gas flow information;

determining at least one group of sample data according to the detection result of the product image; wherein the sample data comprises product attribute information associated with the product image, pre-acquired offset printing speed information, and the gas flow information;

and training a radial basis function neural network by using the sample data to obtain a gas quantity prediction model.

3. The method of claim 2, wherein the product image is a three-dimensional image;

4. The method of claim 3, wherein training a radial basis function neural network with the sample data to obtain a gas quantity prediction model comprises:

5. The method of claim 2, wherein the gas quantity prediction model comprises tuning parameters;

6. The method of claim 5, wherein the optimizing the adjustment parameters by a genetic algorithm until an optimization termination condition is satisfied and outputting an optimization result comprises:

7. The method of claim 6, wherein the adjustment parameters include a center vector and a base width vector;

8. A gas amount prediction device, comprising:

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the gas quantity prediction method of any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions for causing a processor to perform the gas quantity prediction method of any one of claims 1-7 when executed.