CN113858566B - Injection molding machine energy consumption prediction method and system based on machine learning - Google Patents
Injection molding machine energy consumption prediction method and system based on machine learning Download PDFInfo
- Publication number
- CN113858566B CN113858566B CN202111133595.4A CN202111133595A CN113858566B CN 113858566 B CN113858566 B CN 113858566B CN 202111133595 A CN202111133595 A CN 202111133595A CN 113858566 B CN113858566 B CN 113858566B
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- data
- consumption prediction
- parameter
- obtaining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
- B29C45/17—Component parts, details or accessories; Auxiliary operations
- B29C45/76—Measuring, controlling or regulating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76494—Controlled parameter
- B29C2945/76518—Energy, power
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76929—Controlling method
- B29C2945/76979—Using a neural network
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Mechanical Engineering (AREA)
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
The invention discloses an energy consumption prediction method and system for an injection molding machine based on machine learning, wherein the method comprises the following steps: obtaining preset parameter information; acquiring data of an injection molding machine to obtain an original data set; training data and test data are obtained; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data; and inputting the monitoring data serving as input data into the optimized energy consumption prediction model to obtain an output result. The method solves the technical problems that the energy consumption of the injection molding machine cannot be accurately and rapidly predicted due to the defects of low prediction precision and low prediction speed in the injection molding machine energy consumption prediction method in the prior art, and further the working and running conditions of equipment cannot be mastered in real time.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to an energy consumption prediction method and system for an injection molding machine based on machine learning.
Background
The injection molding machine is a key device on a plastic process production line, most of the existing injection molding machines adopt a variable pump or a volume speed regulation hydraulic transmission control mode of the variable pump and a quantitative pump, and the injection molding machine has high running power, serious energy loss and low molding efficiency. Therefore, accurate prediction of energy loss of the injection molding machine is needed, so that the working and running conditions of the equipment can be mastered in real time.
However, in the process of implementing the technical scheme of the invention in the embodiment of the application, the inventor of the application finds that at least the following technical problems exist in the above technology:
in the prior art, the energy consumption prediction method of the injection molding machine has the defects of low prediction precision and low prediction speed, so that the energy consumption of the injection molding machine cannot be accurately and rapidly predicted, and further the working and running conditions of equipment cannot be mastered in real time.
Disclosure of Invention
Aiming at the defects in the prior art, the embodiment of the application aims to solve the technical problems of low prediction precision and low prediction speed in the energy consumption prediction method of the injection molding machine in the prior art by providing the energy consumption prediction method and the energy consumption prediction system of the injection molding machine based on machine learning. The energy consumption of the injection molding machine is predicted by adopting the GRU network model, and simultaneously, the GRU is optimized by combining the attention mechanism and the GSA so as to improve the model precision, so that the energy consumption of the injection molding machine is predicted more accurately, the prediction precision and the prediction speed of the energy consumption prediction model of the injection molding machine are improved, the generalization capability of the model is improved, the energy consumption prediction of the injection molding machine is more accurate, and the molding efficiency of plastic products is improved.
In one aspect, embodiments of the present application provide a machine learning-based energy consumption prediction method for an injection molding machine, wherein the method is applied to an energy consumption prediction system for the injection molding machine, the system including a plurality of industrial sensors, the method comprising: obtaining preset parameter information; according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement; obtaining training data and test data according to the original data set; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information; and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value.
In another aspect, the present application further provides an energy prediction system for an injection molding machine based on machine learning, wherein the system includes: a first obtaining unit: the first obtaining unit is used for obtaining preset parameter information; the first acquisition unit: the first acquisition unit is used for acquiring data of the injection molding machine through the industrial sensor according to the preset parameter information to obtain an original data set, and the original data set has a first time requirement; a second obtaining unit: the second obtaining unit is used for obtaining training data and test data according to the original data set; a first training unit: the first training unit is used for training the AGRU neural network by utilizing the training data based on original parameters, and constructing an energy consumption prediction model; a third obtaining unit: the third obtaining unit is used for obtaining a first moderate value; a first judgment unit: the first judging unit is used for judging whether the first moderate value meets a first preset condition or not; fourth obtaining unit: the fourth obtaining unit is used for obtaining an optimized energy consumption prediction model when the energy consumption prediction model is met, and testing the energy consumption prediction model through the test data; fifth obtaining unit: the fifth obtaining unit is used for obtaining monitoring data through the industrial sensor, and the monitoring data corresponds to the preset parameter information; a first input unit: the first input unit is used for inputting the monitoring data into the optimized energy consumption prediction model as input data to obtain an output result, and the output result comprises an energy consumption prediction value.
One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
obtaining preset parameter information; according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement; obtaining training data and test data according to the original data set; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information; and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value. The energy consumption of the injection molding machine is predicted by adopting the GRU network model, and simultaneously, the GRU is optimized by combining the attention mechanism and the GSA so as to improve the model precision, so that the energy consumption of the injection molding machine is predicted more accurately, the prediction precision and the prediction speed of the energy consumption prediction model of the injection molding machine are improved, the generalization capability of the model is improved, the energy consumption prediction of the injection molding machine is more accurate, and the molding efficiency of plastic products is improved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of an energy consumption prediction method of an injection molding machine based on machine learning according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of an energy prediction method of an injection molding machine based on machine learning when the first moderate value does not meet the first predetermined condition according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of preprocessing the original data set of an energy prediction method of an injection molding machine based on machine learning according to an embodiment of the present application;
FIG. 4 is a schematic flow chart of normalizing the second set of processing parameters according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a machine learning based energy prediction system for an injection molding machine according to an embodiment of the present application;
Fig. 6 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Detailed Description
The embodiment of the application solves the technical problems of low prediction precision and low prediction speed in the energy consumption prediction method of the injection molding machine in the prior art by providing the energy consumption prediction method and the energy consumption prediction system of the injection molding machine based on machine learning. The energy consumption of the injection molding machine is predicted by adopting the GRU network model, and simultaneously, the GRU is optimized by combining the attention mechanism and the GSA so as to improve the model precision, so that the energy consumption of the injection molding machine is predicted more accurately, the prediction precision and the prediction speed of the energy consumption prediction model of the injection molding machine are improved, the generalization capability of the model is improved, the energy consumption prediction of the injection molding machine is more accurate, and the molding efficiency of plastic products is improved.
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Summary of the application
The injection molding machine is a key device on a plastic process production line, most of the existing injection molding machines adopt a variable pump or a volume speed regulation hydraulic transmission control mode of the variable pump and a quantitative pump, and the injection molding machine has high running power, serious energy loss and low molding efficiency. Therefore, accurate prediction of energy loss of the injection molding machine is needed, so that the working and running conditions of the equipment can be mastered in real time. In the prior art, the energy consumption prediction method of the injection molding machine has the defects of low prediction precision and low prediction speed, so that the energy consumption of the injection molding machine cannot be accurately and rapidly predicted, and further the working and running conditions of equipment cannot be mastered in real time.
Aiming at the technical problems, the technical scheme provided by the application has the following overall thought:
the embodiment of the application provides an energy consumption prediction method of an injection molding machine based on machine learning, wherein the method is applied to an energy consumption prediction system of the injection molding machine, the system comprises a plurality of industrial sensors, and the method comprises the following steps: obtaining preset parameter information; according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement; obtaining training data and test data according to the original data set; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information; and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value.
In order to better understand the above technical solutions, the following detailed description will refer to the accompanying drawings and specific embodiments.
Example 1
As shown in fig. 1, an embodiment of the present application provides a machine learning-based energy prediction method for an injection molding machine, wherein the method is applied to an energy prediction system for an injection molding machine, the system includes a plurality of industrial sensors, and the method includes:
step S100: obtaining preset parameter information;
step S200: according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement;
specifically, the injection molding machine is the critical equipment on the plastics technology production line, and present injection molding machine mostly adopts variable pump or variable pump plus quantitative pump's volume speed governing hydraulic drive control mode, and its running power is high, causes the energy loss to lose seriously, and shaping inefficiency is in order to carry out accurate prediction to the energy loss of injection molding machine, is convenient for grasp equipment operation condition in real time, in this application embodiment, has put forward to use GRU network model to predict the energy consumption of injection molding machine, combines attention mechanism and GSA to optimize GRU simultaneously in order to improve model precision, and then makes the energy consumption prediction to the injection molding machine more accurate. Specifically, the preset parameter information includes six main parameters affecting energy consumption in the injection molding machine process: the extrusion speed, extrusion pressure, extrusion temperature, ingot cylinder temperature, blank temperature and die temperature are acquired through six main parameters based on a plurality of industrial sensors, such as a temperature sensor can acquire temperature data of the extrusion temperature, blank temperature and the like, a pressure sensor can acquire pressure data of the extrusion pressure and the like, the plurality of industrial sensors correspond to parameters in preset parameter information, the energy consumption of an injection molding machine is further calculated, the original data set is the parameter of the injection molding machine during operation in history acquired based on six main parameters affecting the energy consumption in the injection molding machine process, and it is noted that the first time requirement needs to be ensured when acquiring data, and the first time requirement can be understood as data acquisition of uniform time, so that the excessive concentration of the data is avoided, and the accuracy of a prediction result is influenced.
More specifically, the GRU neural network is obtained by improving the LSTM neural network, and three gates in the LSTM are reduced to two because the LSTM gating network is too complex and redundant in structure. The GRU performs targeted processing on the time sequence data through an update gate and a reset gate, wherein the update gate determines the retention degree of the previous state information in the current state, and the larger the value is, the more the previous state information is retained. The reset gate is used to determine whether to combine the current state with the previous information, the smaller the value of which indicates the more information is ignored. The model training efficiency is improved. The model is different from the traditional LSTM network in data learning, and the GRU network with higher prediction efficiency can deeply mine the input energy consumption data of the injection molding machine, so that the multidimensional time sequence can establish a nonlinear relation with the energy-consuming time sequence. The influence of data noise is reduced, and the prediction efficiency and accuracy of the whole model are improved.
Step S300: obtaining training data and test data according to the original data set;
step S400: training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model;
Specifically, after the original data set is collected, training data and test data can be further obtained, wherein the test data can be understood as that the original data set is randomly extracted, random uniform distribution of extracted data is ensured, the training data is the rest data set of the test data which is divided in the original data set, for example, if the energy consumption parameter of the injection molding machine in the past year is subjected to traversal collection, data of 5 days are randomly extracted every month as the test data, the rest is used as the training data, further, according to the training data, the AGRU neural network can be trained according to the original parameters of the GSA algorithm and the training data, namely, the collected parameter information such as extrusion speed, extrusion pressure, extrusion temperature, ingot barrel temperature, blank temperature, mold temperature is input into the AGRU neural network, and through continuous training, a predicted value of energy consumption under the parameter information can be obtained, namely, the used electric quantity is predicted by using a GRU with a attention mechanism, the production energy consumption of the injection molding machine, namely, the model is called as the prediction model.
Step S500: obtaining a first moderate value;
step S600: judging whether the first moderate value meets a first preset condition or not;
step S700: when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data;
specifically, the constructed energy consumption prediction model can carry out recursive cyclic analysis operation on the acquired training data, so that the operation obtains the energy consumption predicted value under each parameter, the analysis operation of the fitness function is carried out on the energy consumption predicted value under each parameter, the moderate value corresponding to each parameter can be obtained, the fitness function is specifically called an evaluation function, the fitness function is a standard for distinguishing the quality of individuals in the group, the standard is always nonnegative, and the larger the value of the fitness function is expected to be, the better the value of the fitness function is in any case. The first moderate value is a corresponding fitness value obtained by inputting an energy consumption predicted value of a certain group of parameters into a fitness function to operate, whether the first moderate value meets a first preset condition or not is judged, namely whether the first moderate value is the maximum value or not and whether the first moderate value reaches a termination condition or not is judged, if the first moderate value meets the termination condition, the energy consumption predicted value under the parameters corresponding to the first moderate value is the most accurate, the termination condition of a prediction process is achieved, the corresponding optimal parameter is obtained based on the first moderate value, the optimal energy consumption prediction model is obtained, the energy consumption prediction model is tested through the test data, and the detection accuracy of the optimal energy consumption prediction model is further detected.
Step S800: obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information;
step S900: and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value.
Specifically, the optimized energy consumption prediction model is built according to historical collected data training, in order to ensure detection accuracy and real-time updating performance of the optimized energy consumption prediction model, monitoring data can be obtained through the industrial sensor, the monitoring data correspond to the preset parameter information, namely, based on the industrial sensor, working data corresponding to all parameter values of the injection molding machine during working are collected, the monitoring data are further used as input data to be input into the optimized energy consumption prediction model, an output result is obtained, the output result comprises an energy consumption prediction value, and rapid and accurate prediction of energy consumption of the injection molding machine based on the optimized energy consumption prediction model is achieved.
Further, as shown in fig. 2, after the determining whether the first moderate value meets the first predetermined condition, step S600 further includes:
Step S610: when the first moderate value does not meet the first preset condition, re-selecting parameters to obtain first parameters;
step S620: training an AGRU neural network by using the first parameter, and constructing an energy consumption prediction model;
step S630: obtaining an fitness function;
step S640: performing fitness calculation on the energy consumption prediction model according to the fitness function to obtain a second moderate value;
step S650: judging whether the second moderate value meets the first preset condition or not;
step S660: when the energy consumption prediction model is satisfied, determining a first parameter as an optimal parameter, and determining the optimal energy consumption prediction model according to the first parameter;
step S670: and when the first parameter is not met, repeating the steps, and reselecting the second parameter to obtain a third moderate value until the calculated moderate value meets the first preset condition.
Specifically, when judging whether the first moderate value meets a first preset condition, if the first moderate value does not meet the first preset condition, that is, if the first moderate value is not the maximum value, the parameters can be reselected to obtain first parameters, the first parameters are training data different from the original parameters, further, the first parameters are used for training the AGRU neural network, an energy consumption prediction model is built, training results under the first parameters are obtained, meanwhile, based on the fitness function, fitness calculation is carried out on the training results under the first parameters, a second moderate value is obtained, the second moderate value corresponds to the first parameters one by one, further, whether the second moderate value meets the first preset condition is judged, that is, whether the second moderate value meets the termination condition of the prediction process is met or not is judged, if yes, the first parameters are determined to be optimal parameters, the optimal energy consumption prediction model is determined according to the first parameters, and the energy consumption of the injection molding machine is accurately predicted based on the optimal energy consumption prediction model; otherwise, if the condition is not met, the fact that the current data cannot reach the termination condition of the prediction process is indicated, the steps are continuously repeated, the parameters which are different from the trained parameters are selected again to be retrained until the obtained moderate value meets the first preset condition, the GRU network model dosage form is optimized by the finally obtained moderate value, the optimized energy consumption prediction model is generated, and the energy consumption prediction result of the injection molding machine is more accurate.
It should be noted that when the parameter of the AGRU model is optimized based on the GSA algorithm, the particle speed and the position of the GSA need to be initialized, and the weight of the AGRU is selected at the same time, so that important information is focused from a large amount of information, calculation of unimportant information is reduced, and the aspects of effective selection of information, information correlation and the like are improved. The essence of attention is weighted summation, different input features distribute attention by calculating different probabilities, key features can obtain larger probability values and can be given more attention, and the attention mechanism has better effect when processing time series data. Because the GRU cannot flexibly distinguish which time data in the previous period has a larger influence on multiple targets of the injection molding machine, the attention mechanism provides a means for paying attention to important information, and the capability of the GRU for feature selection in long-time sequence learning is enhanced. When selecting the weight of AGRU, attention weight e is first calculated by a multi-layer perceptron (MLP) t′t I.e. e t′t ＝V T tanh(W s S t′-1 +W h h t ) Wherein e is t′t Representing hidden layer state h t Outputting the attention weight of the influence on the time t' at the time t; v (V) T 、W S And W is h Is the model weight; is h t Encoder hidden layer state; s is S t′-1 Concealing the layer state for the decoder, then pair e according to the softmax function t′t Normalization is carried out to obtain the value of the attention distribution probability distribution, namelyAll weights sum to 1, i.e., Σ i α ti =1, weighted sum to find c t I.e. +.>c t With a connection to the final output of the decoder, the state of the hidden layer of the decoder can be determined by c t An update is performed, this update procedure being simply denoted S t′ ＝GRU(Y t′-1 ,c t′ ,S t′-1 ). The AGRU model is specifically divided into: input layer, hiddenThe hidden layer, the attention layer and the output layer adopt encoder-decoder structure. The input sequence is (X) 1 ,X 2 ,…,X t ) The hidden layer is (h) 1 ,h 2 ,…,h t ) Each input X to the hidden state h is calculated to obtain an output Y of the model at the attention layer by using the attention mechanism theory through the GRU network t′ 。
After determining the weight of AGRU, the weight of AGRU model needs to be optimized based on GSA algorithm. Further, the Gravity Search Algorithm (GSA) is a global optimization algorithm with high robustness and easy implementation, and the weight of the AGRU neural network can be used as the attribute of the particles in the space, and the output error is used as the objective function. The basic principle of the gravity search algorithm can be summarized as: attraction exists between the individual particles, and the magnitude of attraction is proportional to the mass of the particles and inversely proportional to the particle distance. The particles move continuously in the search space by the gravitational force between them, and when the particles move to the optimal position, the optimal solution can be obtained. In the GSA algorithm iteration process, the exploratory capacity and the searching capacity of the algorithm are adjusted by controlling the number of particles in the kbest set. More particles have global attractive force in the early stage of algorithm iteration, so that the diversity of groups is enhanced, and the algorithm has stronger global optimizing capability. At the later stage of algorithm iteration, fewer particles have global attractive force, which is beneficial to the convergence of the whole population, thereby enhancing the local searching capability of the algorithm.
More specifically, when optimizing the weight of the AGRU model based on the GSA algorithm, it is now assumed that the spatial dimension is D, the total number of objects is N, and the position of the ith particle in space is:wherein->Indicating the position of the ith particle in the d-th dimensional space, N is the total population particle number. Mass M of particles i The formula (t) can be deduced as follows: fit i (t)＝f(x i (t)),i＝1,2,…,N；/> The attraction and acceleration to which each particle is subjected are calculated according to the following formula: /> Wherein F is i d (t) and->Representing the force and corresponding acceleration experienced by the particle i in the d-dimension. R is R ij The euclidean distance between particle i and particle j is shown. r is (r) j Is [0,1]The random number epsilon is a minimum number related to the calculation precision. kbest represents the set of K particles that contains the optimal fitness value. The initial value of K is typically the total number of particles in the population and eventually becomes 1 as the iterations of the algorithm decrease linearly. G (t) represents the gravitational constant, which can be expressed by the formula +.>And (3) calculating, wherein G and alpha represent two constants, T is the number of iterations of the current population, and T is the total number of iterations of the algorithm. And further update the velocity v of the particles according to the formula i (t) and position x i (t) the formula: /> Wherein t is the number of iterations of the current population, r i Is [0,1]Random numbers in between.
Further, after the optimized energy consumption prediction model is obtained, step S700 further includes:
Step S710: according to the formula:performing generalization value calculation on the optimized energy consumption prediction model to obtain a first generalization value, wherein m is a natural number and Y is a model output result;
step S720: judging whether the first generalization value meets a second preset threshold value or not;
step S730: and determining the optimized energy consumption prediction model when the first generalization value is smaller than the second preset threshold value.
Step S740: according to the formula:calculating to obtain an error value, wherein m is a natural number, and Y is a model output result;
step S750: judging whether the error value meets a third preset threshold value or not;
step S760: and determining the optimized energy consumption prediction model when the error value is smaller than the third preset threshold value.
Specifically, after the optimized energy consumption prediction model is obtained, in order to further improve the prediction accuracy of the model, RMSE and MAE may be used to evaluate the generalization capability and output error of the model, where RMSE is a root mean square error, which is the square root of the ratio of the square of the deviation between the predicted value and the actual value to the number of observations n, in actual measurement, the number of observations n is always limited, the true value can only be replaced by the most reliable (optimal) value, the root mean square error is used to measure the deviation between the observed value and the true value, MAE is an average absolute error, and average absolute error MAE and root mean square error RMSE are two most common indicators for measuring the variable accuracy, and are two important scales for evaluating the model in machine learning.
The formula may be based on:calculating a generalization value of the optimized energy consumption prediction model, wherein m is a natural number, Y is a model output result, a first generalization value is obtained, and the first generalization value is further judgedWhether a second preset threshold is met or not is judged, the second preset threshold is a preset model standard generalization value, if the first generalization value is smaller than the second preset threshold, the first generalization value is smaller, so that the generalization capability of a corresponding model is stronger, and the optimized energy consumption prediction model is further determined.
Meanwhile, according to the formula:calculating to obtain an error value, wherein m is a natural number, Y is a model output result, judging whether the error value meets a third preset threshold, wherein the third preset threshold can be understood as a preset model standard error value, if the error value is smaller than the third preset threshold, the error value is smaller, so that the prediction accuracy of a corresponding model is higher, and the optimal energy consumption prediction model is further determined.
And the generalization capability and the output error of the model are evaluated by using RMSE and MAE, so that the generalization capability and the prediction accuracy of the optimized energy consumption prediction model are higher.
Further, as shown in fig. 3, before the training data and the test data are obtained according to the original data set, the preprocessing is performed on the original data set, and step S300 further includes:
step S310: obtaining original monitoring data through the industrial sensor, wherein the original monitoring data corresponds to the preset parameter information, and the preset parameter information comprises at least two parameter information;
step S320: obtaining a first parameter according to the preset parameter information;
step S330: grouping the original monitoring data according to the first parameter to obtain monitoring process grouping data;
step S340: and obtaining the original data set according to the monitoring process grouping data.
Specifically, before training data and test data are obtained according to the original data set, the original data set is further required to be preprocessed, further, the original monitoring data can be obtained through the industrial sensor, the original monitoring data can be understood to be historical working parameters of an injection molding machine collected based on parameters such as different extrusion speeds, extrusion pressures, extrusion temperatures, ingot barrel temperatures, blank temperatures, mold temperatures and the like, the original monitoring data are more and more redundant in data types, therefore, the original monitoring data are required to be grouped, namely, data sets corresponding to each type of parameters are regrouped, so that data distribution is clearer and more concentrated, the original monitoring data are grouped according to the first parameter, monitoring process grouping data are obtained, for example, all historical extrusion speed data are contained under extrusion speed grouping, all historical blank temperature data are contained under blank temperature grouping, and the like, the original monitoring process grouping data are obtained, and the obtained original data set is further, and the acquired original monitoring data are classified by parameters, so that massive data are ensured to be managed and clearer and more concentrated.
Further, as shown in fig. 4, the embodiment of the present application further includes:
step S331: classifying the original monitoring data according to the preset parameter information to obtain a parameter classification information set, wherein the parameter classification information set comprises a first parameter set, a second parameter set and an N parameter set, wherein N is a positive integer greater than 2;
step S332: sequentially determining corresponding parameter average values according to the first parameter set, the second parameter set and the N parameter set;
step S333: determining an outlier according to the parameter mean;
step S334: performing outlier elimination on the first parameter set, the second parameter set and the N parameter set according to the outlier to obtain a first processing parameter set;
step S335: obtaining a supplementary parameter information set;
step S336: supplementing the optimized parameter classification set according to the supplementing parameter information set to obtain a second processing parameter set, wherein the quantity of the supplementing parameter information corresponds to the quantity of the eliminating parameters;
step S337: and carrying out normalization processing on the second processing parameter set to obtain the original data set.
Specifically, the data of the injection molding machine come from a variety of industrial sensors. Monitoring data may be lost or anomalous due to sensor failures, network outages, abnormal transmissions, and other problems. The energy consumption and the process parameter change of the injection molding machine can be regarded as normal random processes, in order to extract purer training data based on the original monitoring data, the original monitoring data can be classified according to the preset parameter information to obtain a parameter classification information set, the parameter classification information set comprises a first parameter set, a second parameter set and up to an nth parameter set, in other words, the original monitoring data which is redundant is divided into the first parameter set with data concentrated on the extrusion speed, the second parameter set with data concentrated on the blank temperature, the third parameter set with data concentrated on the mold temperature and the like, until the data in the parameter set corresponding to the nth parameter set is obtained, and then parameter average values corresponding to the parameter sets are obtained, namely, the parameter average value of the extrusion speed is obtained, and further, based on the parameter average value, outlier values are determined, wherein the outlier values can be defined as a group of abnormal values which are more than three times of standard deviation of the average value, and the first processing parameter set can be obtained by rejecting abnormal values, and the first processing parameter set is the original monitoring data after the abnormal values are rejected; meanwhile, a supplementary parameter information set can be obtained, namely, a parameter set different from the preset parameter information is obtained, the optimized parameter classification set is supplemented to obtain a second processing parameter set, the second processing parameter set is obtained by carrying out numerical value complementation on the basis of the first processing parameter set, and then the obtained parameter set is further normalized, so that the convergence speed and precision of a model are improved, the original data set is finally obtained, and the fact that purer training data are extracted based on the original monitoring data is achieved, and further the prediction result of the optimized energy consumption prediction model is more accurate is achieved.
Further, before the supplementing the optimized parameter classification set according to the supplementing parameter information set, step S336 further includes:
step S3361: obtaining a filtering algorithm;
step S3362: and filtering the supplementary parameter information set through the filtering algorithm.
Specifically, before supplementing the optimized parameter classification set, the supplemental parameter information set may be subjected to filtering processing by the filtering algorithm, and further, the filtering processing of the original data mainly removes random errors in the original data to improve the data quality, for example, the filtering algorithm may be a kalman filtering algorithm, which is not only used in a linear system, but also suitable for a nonlinear system, particularly suitable for a system of computer recursion processing, and is used for data preprocessing. It has three functional features: first, historical data can be smoothed; second, current data may be filtered; third, it can predict future data. And finally, carrying out normalization processing to obtain an input data set of the model. The Kalman filter is used for carrying out data smoothing completion, so that the input training data is purer, and the quality of the input training data is improved.
Compared with the prior art, the invention has the following beneficial effects:
1. obtaining preset parameter information; according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement; obtaining training data and test data according to the original data set; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information; and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value. The energy consumption of the injection molding machine is predicted by adopting the GRU network model, and simultaneously, the GRU is optimized by combining the attention mechanism and the GSA so as to improve the model precision, so that the energy consumption of the injection molding machine is predicted more accurately, the prediction precision and the prediction speed of the energy consumption prediction model of the injection molding machine are improved, the generalization capability of the model is improved, the energy consumption prediction of the injection molding machine is more accurate, and the molding efficiency of plastic products is improved.
Example two
Based on the same inventive concept as the injection molding machine energy consumption prediction method based on machine learning in the foregoing embodiment, the present invention further provides an injection molding machine energy consumption prediction system based on machine learning, as shown in fig. 5, the system includes:
the first obtaining unit 11: the first obtaining unit 11 is configured to obtain preset parameter information;
the first acquisition unit 12: the first collection unit 12 is configured to collect data of the injection molding machine through an industrial sensor according to the preset parameter information, so as to obtain an original data set, where the original data set has a first time requirement;
the second obtaining unit 13: the second obtaining unit 13 is configured to obtain training data and test data according to the original data set;
the first training unit 14: the first training unit 14 is configured to train the AGRU neural network by using the training data based on the original parameters, and construct an energy consumption prediction model;
the third obtaining unit 15: the third obtaining unit 15 is configured to obtain a first moderate value;
the first judgment unit 16: the first judging unit 16 is configured to judge whether the first moderate value meets a first predetermined condition;
fourth obtaining unit 17: the fourth obtaining unit 17 is configured to obtain an optimized energy consumption prediction model when the energy consumption prediction model is satisfied, and test the energy consumption prediction model according to the test data;
Fifth obtaining unit 18: the fifth obtaining unit 18 is configured to obtain, by using the industrial sensor, monitoring data, where the monitoring data corresponds to the preset parameter information;
the first input unit 19: the first input unit 19 is configured to input the monitoring data as input data into the optimized energy consumption prediction model, and obtain an output result, where the output result includes an energy consumption prediction value.
Further, the system further comprises:
sixth obtaining unit: the sixth obtaining unit is configured to reselect a parameter to obtain a first parameter when the first moderate value does not satisfy the first predetermined condition;
a second training unit: the second training unit is used for training the AGRU neural network by using the first parameter to construct an energy consumption prediction model;
seventh obtaining unit: the seventh obtaining unit is used for obtaining the fitness function;
a first calculation unit: the first computing unit is used for computing the fitness of the energy consumption prediction model according to the fitness function to obtain a second moderate value;
a second judgment unit: the second judging unit is used for judging whether the second moderate value meets the first preset condition;
A first determination unit: the first determining unit is used for determining a first parameter as an optimal parameter when the first parameter is met, and determining the optimal energy consumption prediction model according to the first parameter;
a first repeat unit: the first repeating unit is configured to repeat the above steps when the first parameter is not satisfied, and reselect the second parameter to obtain a third moderate value until the calculated moderate value satisfies the first predetermined condition.
Further, the system further comprises:
a second calculation unit: the second computing unit is used forAccording to the formula:performing generalization value calculation on the optimized energy consumption prediction model to obtain a first generalization value, wherein m is a natural number and Y is a model output result;
a third judgment unit: the third judging unit is used for judging whether the first generalization value meets a second preset threshold value or not;
a second determination unit: the second determining unit is configured to determine the optimized energy consumption prediction model when the first generalization value is smaller than the second predetermined threshold.
Further, the system further comprises:
a third calculation unit: the third calculation unit is configured to calculate, according to the formula:calculating to obtain an error value, wherein m is a natural number, and Y is a model output result;
Fourth judgment unit: the fourth judging unit is used for judging whether the error value meets a third preset threshold value or not;
a third determination unit: the third determining unit is configured to determine the optimized energy consumption prediction model when the error value is smaller than the third predetermined threshold.
Further, the system further comprises:
eighth obtaining unit: the eighth obtaining unit is configured to obtain, by using the industrial sensor, raw monitoring data, where the raw monitoring data corresponds to the preset parameter information, and the preset parameter information includes at least two parameter information;
a ninth obtaining unit: the ninth obtaining unit is configured to obtain a first parameter according to the preset parameter information;
a first grouping unit: the first grouping unit is used for grouping the original monitoring data according to the first parameter to obtain monitoring process grouping data;
tenth obtaining unit: the tenth obtaining unit is configured to obtain the raw data set according to the monitoring process packet data.
Further, the system further comprises:
a first classification unit: the first classification unit is used for classifying the original monitoring data according to the preset parameter information to obtain a parameter classification information set, wherein the parameter classification information set comprises a first parameter set, a second parameter set and an N parameter set, wherein N is a positive integer greater than 2;
A fourth determination unit: the fourth determining unit is used for determining a corresponding parameter mean value according to the first parameter set, the second parameter set and the N parameter set in sequence;
fifth determining unit: the fifth determining unit is used for determining an outlier according to the parameter mean;
a first rejecting unit: the first eliminating unit is used for eliminating outliers from the first parameter set, the second parameter set and the N parameter set according to the outliers to obtain a first processing parameter set;
eleventh obtaining unit: the eleventh obtaining unit is configured to obtain a supplementary parameter information set;
a first supplementing unit: the first supplementing unit is used for supplementing the optimized parameter classification set according to the supplementing parameter information set to obtain a second processing parameter set, wherein the quantity of the supplementing parameter information corresponds to the quantity of the eliminating parameters;
a twelfth obtaining unit: the twelfth obtaining unit is configured to perform normalization processing on the second processing parameter set, and obtain the original data set.
Further, the system further comprises:
thirteenth obtaining unit: the thirteenth obtaining unit is used for obtaining a filtering algorithm;
A first filtering unit: the first filtering unit is used for carrying out filtering processing on the supplementary parameter information set through the filtering algorithm.
The various variations and embodiments of the machine learning-based energy consumption prediction method for an injection molding machine in the first embodiment of fig. 1 are equally applicable to the machine learning-based energy consumption prediction system for an injection molding machine in this embodiment, and by the foregoing detailed description of the machine learning-based energy consumption prediction method for an injection molding machine, those skilled in the art can clearly know the implementation method of the machine learning-based energy consumption prediction system for an injection molding machine in this embodiment, so that the description is omitted again for brevity.
Example III
An electronic device of an embodiment of the present application is described below with reference to fig. 6.
Fig. 6 illustrates a structural schematic diagram of an electronic device according to an embodiment of the present application.
Based on the inventive concept of a machine learning based energy prediction method for an injection molding machine in the foregoing example embodiments, the present invention further provides a machine learning based energy prediction system for an injection molding machine, on which a computer program is stored, which when executed by a processor implements the steps of any one of the methods of the machine learning based energy prediction system for an injection molding machine described above.
Where in FIG. 6, a bus architecture (represented by bus 300), bus 300 may comprise any number of interconnected buses and bridges, with bus 300 linking together various circuits, including one or more processors, represented by processor 302, and memory, represented by memory 304. Bus 300 may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., as are well known in the art and, therefore, will not be described further herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e. a transceiver, providing a means for communicating with various other systems over a transmission medium. The processor 302 is responsible for managing the bus 300 and general processing, while the memory 304 may be used to store data used by the processor 302 in performing operations.
The embodiment of the application provides an energy consumption prediction method of an injection molding machine based on machine learning, wherein the method is applied to an energy consumption prediction system of the injection molding machine, the system comprises a plurality of industrial sensors, and the method comprises the following steps: obtaining preset parameter information; according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement; obtaining training data and test data according to the original data set; training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model; obtaining a first moderate value; judging whether the first moderate value meets a first preset condition or not; when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data; obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information; and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (9)
1. An energy consumption prediction method for an injection molding machine based on machine learning, wherein the method is applied to an energy consumption prediction system for the injection molding machine, the system comprising a plurality of industrial sensors, the method comprising:
obtaining preset parameter information;
according to the preset parameter information, carrying out data acquisition on the injection molding machine through the industrial sensor to obtain an original data set, wherein the original data set has a first time requirement;
obtaining training data and test data according to the original data set;
training the AGRU neural network by utilizing the training data based on the original parameters, and constructing an energy consumption prediction model;
obtaining a first moderate value;
judging whether the first moderate value meets a first preset condition or not;
when the energy consumption prediction model is satisfied, an optimized energy consumption prediction model is obtained, and the energy consumption prediction model is tested through the test data;
Obtaining monitoring data through the industrial sensor, wherein the monitoring data corresponds to the preset parameter information;
and inputting the monitoring data as input data into the optimized energy consumption prediction model to obtain an output result, wherein the output result comprises an energy consumption predicted value.
2. The method of claim 1, wherein said determining if said first moderate value meets a first predetermined condition comprises:
when the first moderate value does not meet the first preset condition, re-selecting parameters to obtain first parameters;
training an AGRU neural network by using the first parameter, and constructing an energy consumption prediction model;
obtaining an fitness function;
performing fitness calculation on the energy consumption prediction model according to the fitness function to obtain a second moderate value;
judging whether the second moderate value meets the first preset condition or not;
when the energy consumption prediction model is satisfied, determining a first parameter as an optimal parameter, and determining the optimal energy consumption prediction model according to the first parameter;
and when the first parameter is not met, repeating the steps, and reselecting the second parameter to obtain a third moderate value until the calculated moderate value meets the first preset condition.
3. The method of claim 1, wherein after the obtaining the optimized energy consumption prediction model, comprising:
according to the formula:performing generalization value calculation on the optimized energy consumption prediction model to obtain a first generalization value, wherein m is a natural number and Y is a model output result;
judging whether the first generalization value meets a second preset threshold value or not;
and determining the optimized energy consumption prediction model when the first generalization value is smaller than the second preset threshold value.
4. The method of claim 1, wherein after the obtaining the optimized energy consumption prediction model, comprising:
according to the formula:calculating to obtain an error value, wherein m is a natural number, and Y is a model output result;
judging whether the error value meets a third preset threshold value or not;
and determining the optimized energy consumption prediction model when the error value is smaller than the third preset threshold value.
5. The method of claim 1, wherein the preprocessing the original data set before obtaining training data and test data according to the original data set comprises:
obtaining original monitoring data through the industrial sensor, wherein the original monitoring data corresponds to the preset parameter information, and the preset parameter information comprises at least two parameter information;
Obtaining a first parameter according to the preset parameter information;
grouping the original monitoring data according to the first parameter to obtain monitoring process grouping data;
and obtaining the original data set according to the monitoring process grouping data.
6. The method of claim 5, wherein the method comprises:
classifying the original monitoring data according to the preset parameter information to obtain a parameter classification information set, wherein the parameter classification information set comprises a first parameter set, a second parameter set and an N parameter set, wherein N is a positive integer greater than 2;
sequentially determining corresponding parameter average values according to the first parameter set, the second parameter set and the N parameter set;
determining an outlier according to the parameter mean;
performing outlier elimination on the first parameter set, the second parameter set and the N parameter set according to the outlier to obtain a first processing parameter set;
obtaining a supplementary parameter information set;
supplementing the optimized parameter classification set according to the supplementary parameter information set to obtain a second processing parameter set, wherein the quantity of the supplementary parameter information corresponds to the quantity of the eliminating parameters;
And carrying out normalization processing on the second processing parameter set to obtain the original data set.
7. The method of claim 6, wherein said supplementing the optimized parameter classification set according to the supplemental parameter information set comprises, prior to:
obtaining a filtering algorithm;
and filtering the supplementary parameter information set through the filtering algorithm.
8. An energy consumption prediction system for an injection molding machine based on machine learning, wherein the system comprises:
a first obtaining unit: the first obtaining unit is used for obtaining preset parameter information;
the first acquisition unit: the first acquisition unit is used for acquiring data of the injection molding machine through the industrial sensor according to the preset parameter information to obtain an original data set, and the original data set has a first time requirement;
a second obtaining unit: the second obtaining unit is used for obtaining training data and test data according to the original data set;
a first training unit: the first training unit is used for training the AGRU neural network by utilizing the training data based on original parameters, and constructing an energy consumption prediction model;
a third obtaining unit: the third obtaining unit is used for obtaining a first moderate value;
A first judgment unit: the first judging unit is used for judging whether the first moderate value meets a first preset condition or not;
fourth obtaining unit: the fourth obtaining unit is used for obtaining an optimized energy consumption prediction model when the energy consumption prediction model is met, and testing the energy consumption prediction model through the test data;
fifth obtaining unit: the fifth obtaining unit is used for obtaining monitoring data through the industrial sensor, and the monitoring data corresponds to the preset parameter information;
a first input unit: the first input unit is used for inputting the monitoring data into the optimized energy consumption prediction model as input data to obtain an output result, and the output result comprises an energy consumption prediction value.
9. A machine learning based energy prediction system for an injection molding machine, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1-7 when the program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111133595.4A CN113858566B (en) | 2021-09-27 | 2021-09-27 | Injection molding machine energy consumption prediction method and system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111133595.4A CN113858566B (en) | 2021-09-27 | 2021-09-27 | Injection molding machine energy consumption prediction method and system based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113858566A CN113858566A (en) | 2021-12-31 |
| CN113858566B true CN113858566B (en) | 2023-08-08 |
Family
ID=78990852
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111133595.4A Active CN113858566B (en) | 2021-09-27 | 2021-09-27 | Injection molding machine energy consumption prediction method and system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113858566B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114563992B (en) * | 2022-03-01 | 2023-11-21 | 昆山缔微致精密电子有限公司 | Method and system for improving blanking precision of injection mold |
| CN116647819B (en) * | 2023-07-27 | 2023-11-07 | 深圳市中科智联有限公司 | Instrument energy consumption monitoring method and system based on sensor network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102773981A (en) * | 2012-07-16 | 2012-11-14 | 南京航空航天大学 | Implementation method of energy-saving and optimizing system of injection molding machine |
| AU2013224733A1 (en) * | 2009-05-08 | 2013-10-03 | Accenture Global Services Limited | Building energy consumption analysis system |
| CN110866528A (en) * | 2019-10-28 | 2020-03-06 | 腾讯科技（深圳）有限公司 | Model training method, energy consumption use efficiency prediction method, device and medium |
| CN112001486A (en) * | 2020-08-28 | 2020-11-27 | 河北工业大学 | Load decomposition method based on deep learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200104639A1 (en) * | 2018-09-28 | 2020-04-02 | Applied Materials, Inc. | Long short-term memory anomaly detection for multi-sensor equipment monitoring |
-
2021
- 2021-09-27 CN CN202111133595.4A patent/CN113858566B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2013224733A1 (en) * | 2009-05-08 | 2013-10-03 | Accenture Global Services Limited | Building energy consumption analysis system |
| CN102773981A (en) * | 2012-07-16 | 2012-11-14 | 南京航空航天大学 | Implementation method of energy-saving and optimizing system of injection molding machine |
| CN110866528A (en) * | 2019-10-28 | 2020-03-06 | 腾讯科技（深圳）有限公司 | Model training method, energy consumption use efficiency prediction method, device and medium |
| CN112001486A (en) * | 2020-08-28 | 2020-11-27 | 河北工业大学 | Load decomposition method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 宋李俊 ; 潘安大 ; Jing Shi ; 陈猛 ; .基于DE-GPR的数控机床切削能耗预测.重庆理工大学学报(自然科学).2018,(第11期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113858566A (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113858566B (en) | Injection molding machine energy consumption prediction method and system based on machine learning | |
| CN110427654B (en) | Landslide prediction model construction method and system based on sensitive state | |
| CN113486578A (en) | Method for predicting residual life of equipment in industrial process | |
| CN113188794B (en) | Gearbox fault diagnosis method and device based on improved PSO-BP neural network | |
| CN111122162B (en) | Industrial system fault detection method based on Euclidean distance multi-scale fuzzy sample entropy | |
| CN110880369A (en) | Gas marker detection method based on radial basis function neural network and application | |
| CN111768000A (en) | Industrial process data modeling method for online adaptive fine-tuning deep learning | |
| CN117113236B (en) | Smart city monitoring system and data processing method | |
| CN113887342A (en) | Equipment fault diagnosis method based on multi-source signals and deep learning | |
| CN117390591B (en) | Operation monitoring method and system for coal conveying belt sampling machine based on electric parameter analysis | |
| CN115062272A (en) | Water quality monitoring data abnormity identification and early warning method | |
| CN114015825A (en) | Method for monitoring abnormal state of blast furnace heat load based on attention mechanism | |
| CN115859077A (en) | Multi-feature fusion motor small sample fault diagnosis method under variable working conditions | |
| CN114091525A (en) | Rolling bearing degradation trend prediction method | |
| CN116245227A (en) | Daily weather drought prediction method, device, storage medium and equipment | |
| CN114863170A (en) | Deep learning-based new energy vehicle battery spontaneous combustion early warning method and device | |
| CN117748481A (en) | Real-time dynamic partitioning-based power system inertia online evaluation method and device | |
| CN116821697A (en) | Mechanical equipment fault diagnosis method based on small sample learning | |
| CN117041972A (en) | Channel-space-time attention self-coding based anomaly detection method for vehicle networking sensor | |
| CN115034504B (en) | Cutter wear state prediction system and method based on cloud edge cooperative training | |
| CN116720743A (en) | Carbon emission measuring and calculating method based on data clustering and machine learning | |
| CN116720079A (en) | Wind driven generator fault mode identification method and system based on multi-feature fusion | |
| CN113537459B (en) | Drug warehouse temperature and humidity prediction method | |
| CN114065335A (en) | Building energy consumption prediction method based on multi-scale convolution cyclic neural network | |
| CN113973403A (en) | Temperature-induced strain field redistribution intelligent sensing method based on structure discrete measuring point topology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |