Production scheduling method and system of prefabricated data center module based on deep learning
By applying deep learning-based production scheduling methods and systems in prefabricated modular data center production, various challenges in production scheduling are solved, efficient and precise production planning optimization is achieved, and production efficiency and management level are improved.
Patent Information
- Application Number
- CN202411575336.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-06
- Publication Date
- 2025-02-25
AI Technical Summary
During the factory production process of prefabricated modular data centers, there are problems with equipment and accessories supply, integrated installation resource coordination, production process rhythm problems, equipment failures and maintenance problems, quality control problems, and changes in market and customer demand, resulting in challenges in production scheduling.
Using deep learning-based production scheduling methods and systems, the target deep learning model is determined by collecting and preprocessing the basic data of production scheduling, and training the model based on the preprocessed data to obtain the predicted model. This prediction model is used to generate prediction results and is used to formulate or optimize production plans. The system includes a data collection and processing system, a production scheduling training and optimization system, and a production monitoring and scheduling system.
It realizes accurate and efficient management of production scheduling of prefabricated data center modules, which can optimize production planning, improve production efficiency, reduce waiting time, reduce energy consumption and cost, and improve response speed and management level.
Smart Images
Figure CN119514964A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of production and manufacturing of prefabricated data center modules, and in particular to a production scheduling method and system for prefabricated data center modules based on deep learning. Background Art
[0002] With the rapid development of technologies such as cloud computing, big data and artificial intelligence, the demand for computing power has exploded. With the rapid development of the computing power industry, the construction of traditional data center infrastructure faces problems such as slow speed and difficulty in expansion. These problems urgently need to be solved through technological innovation and model optimization.
[0003] In response to the above problems, engineering productization and product modularization have become the key to solving the difficulties in data center construction. Specifically, by moving the installation process forward to the factory and adopting a prefabricated fully modular data center solution, the construction efficiency and quality of the data center can be greatly improved, including integrating the "wind, fire, water and electricity" equipment in the data center into IT modules, power supply modules, refrigeration modules, etc. These modules are independent and can be integrated. The module construction, equipment installation and factory commissioning are completed in the factory to ensure the quality and performance of the module and reduce the amount of on-site construction. Compared with traditional data center construction, the construction period of a prefabricated data center can be shortened by 50%, greatly improving the delivery speed.
[0004] However, in the process of factory production of prefabricated modular data centers, the production scheduling of prefabricated modular data centers is challenged by the supply of equipment, accessories and other materials, the coordination of integrated installation resources, the rhythm of production processes, equipment failure and maintenance, quality control, and changes in market and customer demand. Summary of the invention
[0005] In view of this, an object of the present invention is to provide a production scheduling method and system for prefabricated data center modules based on deep learning, so as to solve the problem of production scheduling of prefabricated modular data centers.
[0006] To achieve the above objectives, the present invention adopts the following technical solutions:
[0007] In a first aspect, an embodiment of the present application provides a production scheduling method for prefabricated data center modules based on deep learning, comprising:
[0008] Collecting basic data for production scheduling of prefabricated data center modules and preprocessing the basic data;
[0009] Determining a target deep learning model based on the characteristic requirements of the production scheduling of the prefabricated data center module;
[0010] Based on the preprocessed basic data, the target deep learning model is trained to obtain a prediction model;
[0011] A prediction result is generated by the prediction model, and the prediction result is used to formulate or optimize the production plan of the prefabricated data center module.
[0012] Furthermore, in some embodiments of the present application, the basic data includes: production progress, equipment status, material supply and personnel arrangement; the preprocessing includes data cleaning, data conversion, data normalization and feature extraction.
[0013] Furthermore, in some embodiments of the present application, the target deep learning model includes:
[0014] Recurrent Neural Networks, Convolutional Neural Networks, Long Short-Term Memory Networks, and Gated Recurrent Units;
[0015] The recurrent neural network is used to process and analyze the continuous operation of equipment installation in the production process; the convolutional neural network is used to process and analyze data with spatial correlation, wherein the data with spatial correlation includes equipment layout and material transportation data;
[0016] The LSTM network and gated recurrent unit are used to capture long-term dependencies in production scheduling.
[0017] Furthermore, in some embodiments of the present application, it also includes:
[0018] Monitor the execution of production scheduling in real time, monitor the status of each process in the process, and obtain monitoring data;
[0019] The monitoring data is used to improve and optimize the prediction model.
[0020] Furthermore, in some embodiments of the present application, the target deep learning model is trained based on the preprocessed basic data to obtain a prediction model, including:
[0021] During the training process, a gradient descent algorithm and a minimization loss function are used to improve the generalization ability of the prediction model;
[0022] and guiding the prediction model to make optimized scheduling decisions through a reward mechanism; wherein the reward mechanism is generated based on the goals of completing orders, reducing waiting time, and reducing energy consumption.
[0023] In a second aspect, an embodiment of the present application provides a production scheduling system for prefabricated data center modules based on deep learning, characterized in that it includes: a data collection and processing system, a production scheduling training optimization system, and a production monitoring scheduling system;
[0024] The data collection and processing system is used to collect basic data of production scheduling of prefabricated data center modules and preprocess the basic data;
[0025] The production scheduling training optimization system is used to determine the target deep learning model based on the characteristic requirements of the production scheduling of the prefabricated data center module; and train the target deep learning model based on the preprocessed basic data to obtain a prediction model;
[0026] A production monitoring and scheduling system is used to generate prediction results through the prediction model, and the prediction results are used to formulate or optimize the production plan of the prefabricated data center module.
[0027] The present invention relates to the technical field of production and manufacturing of prefabricated data center modules, and in particular to a production scheduling method and system for prefabricated data center modules based on deep learning, the method comprising: collecting basic data for production scheduling of prefabricated data center modules, and preprocessing the basic data; determining a target deep learning model based on the characteristic requirements of production scheduling of the prefabricated data center modules; training the target deep learning model based on the preprocessed basic data to obtain a prediction model; generating a prediction result through the prediction model, and the prediction result is used to formulate or optimize the production plan of the prefabricated data center modules. In this way, the production scheduling of prefabricated data center modules can be accurately and efficiently realized. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0029] Figure 1 is a schematic diagram of a production scheduling method for prefabricated data center modules based on deep learning provided by an embodiment of the present invention;
[0030] Figure 2 is a schematic diagram of a production scheduling system for prefabricated data center modules based on deep learning provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0031] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.
[0032] Figure 1 is a schematic diagram of a production scheduling method for prefabricated data center modules based on deep learning provided by an embodiment of the present invention, Figure 2 is a schematic diagram of a production scheduling system for prefabricated data center modules based on deep learning provided by an embodiment of the present invention (the production scheduling system for prefabricated data center modules based on deep learning provided by an embodiment of the present invention is used to implement the production scheduling method for prefabricated data center modules based on deep learning provided by an embodiment of the present invention), see Figure 1 and Figure 2 The production scheduling method for prefabricated data center modules based on deep learning provided in this embodiment may include the following steps:
[0033] S101. Collect basic data of production scheduling of prefabricated data center modules and pre-process the basic data.
[0034] S102. Determine a target deep learning model based on the characteristic requirements of production scheduling of prefabricated data center modules.
[0035] S103: Based on the preprocessed basic data, the target deep learning model is trained to obtain a prediction model.
[0036] S104. Generate prediction results through the prediction model, and use the prediction results to formulate or optimize the production plan of the prefabricated data center modules.
[0037] The production scheduling method of prefabricated data center modules based on deep learning provided by the present application collects basic data of production scheduling of prefabricated data center modules and preprocesses the basic data; determines the target deep learning model based on the characteristic requirements of production scheduling of prefabricated data center modules; trains the target deep learning model based on the preprocessed basic data to obtain a prediction model; generates prediction results through the prediction model, and the prediction results are used to formulate or optimize the production plan of prefabricated data center modules. In this way, the production scheduling of prefabricated data center modules can be accurately and efficiently realized.
[0038] In some embodiments of the present application, the data collection and processing system can be divided into two parts, namely, a data collection system and a data processing system. The data collection system refers to the software and equipment for collecting and processing the data related to the production scheduling of the prefabricated data center module, namely the basic data, wherein the basic data includes production progress, equipment status, material supply, personnel arrangement, etc.; the data processing system pre-processes the collected basic data, including data cleaning, data conversion, data normalization, feature extraction, etc., in order to meet the input requirements of the deep learning model.
[0039] Furthermore, in some embodiments of the present application, the production scheduling training optimization system refers to selecting a target deep learning model from preset deep learning models according to the characteristics and requirements of production scheduling, training and optimizing the above-mentioned preprocessed data, that is, training and optimizing to obtain a prediction model.
[0040] Among them, the preset deep learning models include: recurrent neural network, convolutional neural network, long short-term memory network and gated recurrent unit. In this application, a recurrent neural network (RNN) can be used to process continuous operations for equipment installation in the production process. A convolutional neural network (CNN) is used to process data with spatial correlation such as equipment layout and material transportation. A long short-term memory network (LSTM) or a gated recurrent unit (GRU) is used to capture long-term dependencies in production scheduling.
[0041] On this basis, through the production monitoring and scheduling system, the trained deep learning model, that is, the prediction model, is applied to the actual production scheduling, so as to formulate and optimize the production plan according to the prediction results of the model. It can also monitor the execution of production scheduling in real time, monitor the status of each process in the process, and feed back the new data collected during the monitoring process to the prediction model, so as to continuously improve and optimize the prediction model.
[0042] In some embodiments of the present application, in the above-mentioned training prediction model process, an optimization method such as a gradient descent algorithm can be used to minimize the loss function and improve the generalization ability of the prediction model. In addition, a reward mechanism can be used to guide the deep learning model to optimize scheduling decisions. For example, the goals of completing orders, reducing waiting time, and reducing energy consumption can be set, and the model learning can be guided by a reward mechanism.
[0043] The production scheduling method for prefabricated data center modules provided in this application, after training the prediction model, applies the trained prediction model, i.e., the deep learning model, to the actual production scheduling, and can formulate and optimize the production plan according to the prediction results of the model. At the same time, by real-time monitoring of the execution of the production scheduling, collecting new data and feeding it back to the model, the model can be continuously improved and optimized.
[0044] The following is a complete example to introduce and explain the production scheduling method of prefabricated data center modules based on deep learning provided in the embodiment of the present application:
[0045] First, the data collection equipment in the data collection and processing system is used to collect data related to the production of prefabricated modular data centers, including the operating status data of AMR, intelligent warehouse, AGV, and server, as well as data on production progress, material supply, and personnel arrangements.
[0046] Then, the data collected by the data acquisition system is preprocessed through the data processing system in the data collection and processing system, including data cleaning, data conversion, data normalization, feature extraction, etc. By preprocessing missing values, abnormal values, duplicate values, and irrelevant values, the data type and format are converted, and then the data is scaled and normalized according to the features. The dimension and complexity of the data are reduced through feature selection, feature extraction, data compression and other methods to improve the training efficiency and performance of the model.
[0047] Then, according to the characteristics and needs of production scheduling, the deep learning model is selected through the production scheduling training optimization system. Among them, for the continuous operation of equipment installation in the production process, the sequence data with time dependence is processed by recurrent neural network (RNN); the data with spatial correlation such as equipment layout and material transportation is processed by convolutional neural network (CNN); the long short-term memory network (LSTM) or gated recurrent unit (GRU) is used to capture the long-term dependencies in production scheduling, predict long-term resource demand and production capacity, and help production decision makers formulate long-term production plans.
[0048] It should be noted that, compared with actual supervision data, the historical data training set that conforms to the data types introduced above can be used to train the deep learning model, and the parameters and structure of the model can be adjusted to enable it to accurately predict the results of production scheduling. The production scheduling model can also be verified using the validation set to select the production scheduling model configuration with the best performance.
[0049] Among them, during the training process, optimization methods such as gradient descent algorithm can be used to minimize the loss function and improve the prediction accuracy and performance of the production scheduling model.
[0050] In this way, the trained deep learning model is applied to the actual production scheduling of prefabricated modular data centers, and the production plan is formulated and optimized based on the prediction results of the model. In addition, the production scheduling execution of prefabricated modular data centers is monitored in real time, and new data is collected and fed back to the model, so as to continuously improve and optimize the model in terms of improving production efficiency, optimizing resource allocation, ensuring product quality, reducing energy consumption and costs, improving response speed, enhancing risk response capabilities, and improving management levels.
[0051] Among them, a reward mechanism can be set up to guide the deep learning model to optimize scheduling decisions. For example, the goals of minimizing production time, maximizing equipment utilization, reducing energy consumption, minimizing waiting time, etc. can be set, and the model learning can be guided by the reward mechanism.
[0052] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.
[0053] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.
[0054] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.
[0055] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0056] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0057] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0058] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.
[0059] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0060] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.
Claims
1. A production scheduling method for prefabricated data center modules based on deep learning, characterized in that: include: Collecting basic data for production scheduling of prefabricated data center modules and preprocessing the basic data; Determining a target deep learning model based on the characteristic requirements of the production scheduling of the prefabricated data center module; Based on the preprocessed basic data, the target deep learning model is trained to obtain a prediction model; A prediction result is generated by the prediction model, and the prediction result is used to formulate or optimize the production plan of the prefabricated data center module.
2. According to the method for production scheduling of prefabricated data center modules based on deep learning in claim 1, it is characterized in that: The basic data includes: production progress, equipment status, material supply and personnel arrangement; the preprocessing includes data cleaning, data conversion, data normalization and feature extraction.
3. The production scheduling method of prefabricated data center modules based on deep learning according to claim 1, characterized in that: The target deep learning model includes: Recurrent Neural Networks, Convolutional Neural Networks, Long Short-Term Memory Networks, and Gated Recurrent Units; The recurrent neural network is used to process and analyze the continuous operation of equipment installation in the production process; the convolutional neural network is used to process and analyze data with spatial correlation, wherein the data with spatial correlation includes equipment layout and material transportation data; The LSTM network and gated recurrent unit are used to capture long-term dependencies in production scheduling.
4. The production scheduling method of prefabricated data center modules based on deep learning according to claim 1, characterized in that: Also includes: Monitor the execution of production scheduling in real time, monitor the status of each process in the process, and obtain monitoring data; The monitoring data is used to improve and optimize the prediction model.
5. The production scheduling method of prefabricated data center modules based on deep learning according to claim 1, characterized in that: The target deep learning model is trained based on the preprocessed basic data to obtain a prediction model, including: During the training process, a gradient descent algorithm and a minimization loss function are used to improve the generalization ability of the prediction model; and guiding the prediction model to make optimized scheduling decisions through a reward mechanism; wherein the reward mechanism is generated based on the goals of completing orders, reducing waiting time, and reducing energy consumption.
6. A production scheduling system for prefabricated data center modules based on deep learning, characterized in that: include: Data collection and processing system, production scheduling training optimization system and production monitoring and scheduling system; The data collection and processing system is used to collect basic data of production scheduling of prefabricated data center modules and preprocess the basic data; The production scheduling training optimization system is used to determine a target deep learning model based on the characteristic requirements of the production scheduling of the prefabricated data center module; Based on the preprocessed basic data, the target deep learning model is trained to obtain a prediction model; A production monitoring and scheduling system is used to generate prediction results through the prediction model, and the prediction results are used to formulate or optimize the production plan of the prefabricated data center module.