A method and system for constructing industrial data hybrid model for discrete manufacturing
Patent Information
- Application Number
- CN202411435720.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-10-15
Smart Images

Figure CN118941174B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent reasoning, and in particular to a method and system for constructing an industrial data hybrid model for discrete manufacturing. Background Art
[0002] As the discrete manufacturing industry enters the era of digital transformation, data-driven intelligent reasoning becomes the key to improving production efficiency and quality and reducing energy consumption. In the discrete manufacturing industry, data may come from multiple sources, including sensors, equipment, production lines, etc., and in the manufacturing environment, many decisions require real-time response, not just analysis of historical data. Conditions in industrial environments may also be constantly changing, requiring reasoning systems to continuously optimize performance in a changing environment to adapt to new processes and requirements. Manufacturing often involves complex production processes involving multiple links and stages. Therefore, discrete manufacturing industrial data has the characteristics of multi-source heterogeneity, massive, high-dimensional, multi-scale, multi-time and space, and high real-time requirements. It is very challenging to establish accurate models to reason about these complex processes.
[0003] Traditional methods of data intelligence reasoning are generally static and simple, and are difficult to flexibly adapt to dynamically changing environments and problems. For application scenarios with high real-time requirements, traditional methods may not be flexible enough. They perform well for a single type of industrial data, but if the data source is multi-source data or multi-time, space and multi-scale data, it is necessary to constantly manually update the rules and logic or select a more appropriate model, which makes data reasoning based on traditional models not flexible enough. Summary of the invention
[0004] Purpose of the invention: The purpose of the present invention is to provide a hybrid model construction method and system based on reinforcement learning, which is used for intelligent reasoning of discrete manufacturing industrial data to solve the problems of multi-source, heterogeneous and multi-scale discrete manufacturing industrial data.
[0005] Technical solution: The method for constructing an industrial data hybrid model for discrete manufacturing according to the present invention comprises the following steps:
[0006] Collect discrete manufacturing industrial data in real time and extract features to obtain feature vectors;
[0007] Passing the characteristic data through the first model and the second model respectively to obtain a first output and a second output;
[0008] Taking a weighted sum of the first output and the second output to obtain a hybrid model of industrial data;
[0009] The first model is used to obtain a fitting curve of the feature data in a discrete space, and the second model is used to perform integrated learning on the feature data using a decision tree as a weak learner.
[0010] Furthermore, the first model calculates the output for each input feature vector ,in Represents industrial losses, is the weight vector of the input vector, is a high-dimensional feature space, is the input feature vector, ;
[0011] By minimizing Solved and , the constraints are met during the solution process:
[0012] ;
[0013] in, is the penalty item, is the actual target value of the sample, is the weight vector, is the threshold of tolerable error, is the slack variable.
[0014] Furthermore, the second model calculates the output for each input feature vector , , where K is the number of weak learners, R is the original feature space of discrete manufacturing industry data, is the input feature vector, ; is a function in R, representing the training K A model of a tree;
[0015] The objective function of the second model is , by iteratively adding tree models to minimize the objective function; where loss is the discrete manufacturing industry dataset loss, M and is the number of decision tree child nodes and their weights, C is the penalty coefficient.
[0016] Furthermore, the hybrid model is ,in is the first output, is the weight of the first output, For the second output, is the weight of the second output.
[0017] Further, the weight of the first output is calculated using the least squares weight fusion method and the weight of the second output .
[0018] Furthermore, the R2 index is used to evaluate the parameters of the hybrid model.
[0019] The discrete manufacturing-oriented industrial data hybrid model construction system of the present invention comprises:
[0020] A feature extraction unit is used to collect discrete manufacturing industrial data in real time and extract features to obtain feature vectors;
[0021] A data processing unit, used for obtaining a first output and a second output by respectively passing the feature data through a first model and a second model; the first model is used for obtaining a fitting curve of the feature data in a discrete space, and the second model is used for performing integrated learning on the feature data using a decision tree as a weak learner;
[0022] The hybrid model building unit is used to perform weighted summation of the first output and the second output to obtain a hybrid model of industrial data.
[0023] Furthermore, the first model calculates the output for each input feature vector ,in Represents industrial losses, is the weight vector of the input vector, is a high-dimensional feature space, is the input feature vector, ;
[0024] By minimizing Solved and , the constraints are met during the solution process:
[0025] ;
[0026] in, is the penalty item, is the actual target value of the sample, is the weight vector, is the threshold of tolerable error, is the slack variable;
[0027] The second model calculates the output for each input feature vector , , where K is the number of weak learners, R is the original feature space of discrete manufacturing industry data, is the input feature vector, ; is a function in R, representing the training K A model of a tree;
[0028] The objective function of the second model is , by iteratively adding tree models to minimize the objective function; where loss is the discrete manufacturing industry dataset loss, M and is the number of decision tree child nodes and their weights, C is the penalty coefficient.
[0029] The electronic device described in the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and is characterized in that when the computer program is loaded into the processor, the method for constructing an industrial data hybrid model for discrete manufacturing is implemented.
[0030] The computer-readable storage medium of the present invention stores a computer program, and when the computer program is executed by a processor, the method for constructing an industrial data hybrid model for discrete manufacturing is implemented.
[0031] Beneficial effect: Compared with the prior art, the advantages of the present invention are: The present invention proposes to pass discrete manufacturing industrial data through two models respectively, apply the least squares weighted fusion method, and output a hybrid model of intelligent reasoning of discrete manufacturing industrial data. It intersects with traditional data intelligent reasoning methods, is more adaptable to dynamically changing environments and problems, and better meets the complex and changeable business scenario requirements of the discrete manufacturing industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 This is a flow chart of a hybrid model construction method according to an embodiment of the present invention. DETAILED DESCRIPTION
[0033] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.
[0034] like Figure 1 As shown, the method for constructing an industrial data hybrid model for discrete manufacturing includes the following steps.
[0035] S1, real-time collection of discrete manufacturing industrial data (including real-time equipment status data, energy consumption data, life cycle data, sensor data, production process data and quality data, etc.). The collected data is mapped from the original feature space to the high-dimensional feature space for easy fitting.
[0036] S2, the discrete manufacturing industrial data are respectively processed by discrete data vector regression (DSVR) algorithm and discrete data gradient boosting (DXGB) algorithm.
[0037] (1) DSVR algorithm:
[0038] By finding a curve in the discrete space and making all data points as close to the curve as possible, the loss is minimized for subsequent data reasoning.
[0039] ,in Represents industrial losses, is the weight vector of the input vector, is a high-dimensional feature space, is the input vector.
[0040] The purpose of DSVR is to minimize the loss To optimize the model, we can solve and , but in the solution process, certain constraints must be observed , It cannot be set too large or too small, so set the variable for each sample of DSVR , to avoid loss of plausibility:
[0041] ;
[0042] Among them, B is the penalty coefficient, which is generally set to 1. is the actual target value of the sample, is the weight vector, is the threshold of tolerable error, is the slack variable.
[0043] (2) DXGB algorithm:
[0044] Taking decision tree as weak learner, the integration mode of learner is: , , For the i The input vector of discrete manufacturing industry data, K is the number of weak learners, R is the original feature space of discrete manufacturing industry data, is a function in R, representing the training K The DXGB objective function is , by iteratively adding tree models to minimize the objective function. loss For discrete manufacturing industry dataset loss, M and is the number of child nodes and their weights, and C is the penalty coefficient.
[0045] S3, after the discrete manufacturing industry data passes through the two models, the least squares weight fusion method is applied to obtain the combined weights, and the output is the hybrid model: ,in is the weight of DSVR in the hybrid model, is the weight of DXGB in the hybrid model, and its loss function is , satisfying the constraints , optimally solve the comprehensive weights so that the loss function Minimum, that is , and get and .in is the weight of DSVR in the hybrid model, is the weight of DXGB in the hybrid model.
[0046] S4, the R2 index is used to evaluate the model parameters.
[0047] ;
[0048] in, is the variance, and MSE is the mean square error. The closer the R2 score is to 1, the better the hybrid model effect is.
[0049] The discrete manufacturing-oriented industrial data hybrid model construction system of the present invention comprises:
[0050] A feature extraction unit is used to collect discrete manufacturing industrial data in real time and extract features to obtain feature vectors;
[0051] A data processing unit, used for obtaining a first output and a second output by respectively passing the feature data through a first model and a second model; the first model is used for obtaining a fitting curve of the feature data in a discrete space, and the second model is used for performing integrated learning on the feature data using a decision tree as a weak learner;
[0052] The hybrid model building unit is used to perform weighted summation of the first output and the second output to obtain a hybrid model of industrial data.
[0053] The electronic device described in the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and is characterized in that when the computer program is loaded into the processor, the method for constructing an industrial data hybrid model for discrete manufacturing is implemented.
[0054] The computer-readable storage medium of the present invention stores a computer program, and when the computer program is executed by a processor, the method for constructing an industrial data hybrid model for discrete manufacturing is implemented.
[0055] The computer-readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, or any other medium that can be used to store the desired program code in the form of instructions or data structures and that can be accessed by a computer.
[0056] The processor is used to execute the computer program stored in the memory to implement each step of the method involved in the above embodiment.
Claims
1. A method for constructing an industrial data hybrid model for discrete manufacturing, characterized in that: The steps include: Collect discrete manufacturing industrial data in real time and extract features to obtain feature vectors; Passing the characteristic data through the first model and the second model respectively to obtain a first output and a second output; Taking a weighted sum of the first output and the second output to obtain a hybrid model of industrial data; The first model is used to obtain a fitting curve of the feature data in a discrete space, and the second model is used to perform integrated learning on the feature data using a decision tree as a weak learner; The first model calculates the output for each input feature vector ,in Represents industrial losses, is the weight vector of the input vector, is a high-dimensional feature space, is the input feature vector, ; ; in, is the penalty item, is the actual target value of the sample, is the weight vector, is the threshold of tolerable error, is the slack variable; The second model calculates the output for each input feature vector , , where K is the number of weak learners, R is the original feature space of discrete manufacturing industry data, is the input feature vector, ; is a function in R, representing the training K A model of a tree; The objective function of the second model is , by iteratively adding tree models to minimize the objective function; where loss is the discrete manufacturing industry dataset loss, M and is the number of decision tree child nodes and their weights, C is the penalty coefficient.
2. The method for constructing an industrial data hybrid model for discrete manufacturing according to claim 1, characterized in that: The hybrid model is ,in is the first output, is the weight of the first output, For the second output, is the weight of the second output.
3. The method for constructing an industrial data hybrid model for discrete manufacturing according to claim 2, characterized in that: The weight of the first output is calculated using the least squares weight fusion method and the weight of the second output .
4. The method for constructing an industrial data hybrid model for discrete manufacturing according to claim 1, characterized in that: The R2 index is used to evaluate the parameters of the hybrid model.
5. An industrial data hybrid model building system for discrete manufacturing, characterized in that: include: A feature extraction unit is used to collect discrete manufacturing industrial data in real time and extract features to obtain feature vectors; A data processing unit, used for obtaining a first output and a second output by passing the feature data through a first model and a second model respectively; the first model is used for obtaining a fitting curve of the feature data in a discrete space, and the second model is used for performing integrated learning on the feature data using a decision tree as a weak learner; A hybrid model building unit, configured to perform a weighted summation of the first output and the second output to obtain a hybrid model of industrial data; The first model calculates the output for each input feature vector ,in Represents industrial losses, is the weight vector of the input vector, is a high-dimensional feature space, is the input feature vector, ; By minimizing Solved and , the constraints are met during the solution process: ; in, is the penalty item, is the actual target value of the sample, is the weight vector, is the threshold of tolerable error, is the slack variable; The second model calculates the output for each input feature vector , , where K is the number of weak learners, R is the original feature space of discrete manufacturing industry data, is the input feature vector, ; is a function in R, representing the training K A model of a tree; The objective function of the second model is , by iteratively adding tree models to minimize the objective function; where loss is the discrete manufacturing industry dataset loss, M and is the number of decision tree child nodes and their weights, C is the penalty coefficient.
6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into a processor, the method for constructing an industrial data hybrid model for discrete manufacturing according to any one of claims 1 to 4 is implemented.
7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for constructing an industrial data hybrid model for discrete manufacturing according to any one of claims 1 to 4 is implemented.
