Digital factory energy consumption management system, management method, medium and computer
By designing a digital factory energy consumption management system, the problem of not being able to automatically monitor and predict factory energy consumption in the existing technology is solved, and automatic recording, accurate prediction and timely alarm of energy consumption are realized, reducing energy costs and production risks.
Patent Information
- Application Number
- CN202411826500.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-03-25
AI Technical Summary
The lack of systems in the prior art that can automatically monitor and predict plant energy consumption, resulting in inaccurate monitoring of energy consumption data and inaccurate prediction.
A digital factory energy consumption management system is designed, including data acquisition, transmission, processing, storage, analysis, prediction, alarm and user management modules. The system collects equipment energy consumption data in real time, performs data preprocessing and storage, and uses machine learning algorithms (such as CNN-LSTM model) to predict and abnormal alarms.
Automatic recording and accurate prediction of factory energy consumption is achieved, and energy consumption alarms are issued in a timely manner to avoid affecting production work due to excessive short-term energy consumption demand and reduce energy costs.
Smart Images
Figure CN119690014A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data management, and more specifically, to a digital factory energy consumption management system and a management method. Background Art
[0002] In the existing technology, an intelligent factory refers to a factory that improves the production efficiency, flexibility and overall management level of the factory by introducing advanced information technology, automation technology, Internet of Things, big data, artificial intelligence, etc. on the basis of traditional manufacturing. The intelligent factory aims to achieve a highly automated, intelligent and digital production process, enabling the factory to self-perceive, self-learn, self-determine and self-optimize.
[0003] In the prior art, there is no system that can automatically monitor, count and predict the energy consumption data of a factory, so as to accurately monitor and accurately predict the energy consumption data of a factory. Summary of the invention
[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a digital factory energy consumption management system, management method, medium and computer to overcome the shortcomings of the prior art that the energy consumption of the factory cannot be automatically monitored and predicted.
[0006] The above technical objectives of the present invention are achieved through the following technical solutions:
[0007] A digital factory energy consumption management system, comprising:
[0008] Data collection module: used to collect real-time energy consumption data of equipment in each area;
[0009] Data transmission module: used to upload the collected real-time energy consumption data to the central system to ensure reliable data transmission and centralized storage;
[0010] Data processing module: used to pre-process the real-time energy consumption data received by the central system and remove abnormal data;
[0011] Data storage module: used to transfer the pre-processed real-time energy consumption data to the database for persistent storage to obtain historical energy consumption data;
[0012] Energy consumption analysis module: used to analyze the historical energy consumption data in the database and generate energy consumption reports;
[0013] Energy consumption prediction module: used to predict the data of the future scheduled stage based on the historical energy consumption data to obtain the energy consumption prediction results;
[0014] Energy consumption alarm module: used for sending an alarm signal when the energy consumption prediction result exceeds the threshold;
[0015] User management module: used to manage system user permissions and ensure system security and access control.
[0016] A digital factory energy consumption management method is applied to the digital factory energy consumption management system of the first aspect, and the management method specifically includes: using the data acquisition module to collect real-time energy consumption data of equipment in each area; using the data transmission module to upload the collected real-time energy consumption data to the central system to ensure reliable transmission and centralized storage of data; using the data processing module to pre-process the real-time energy consumption data received by the central system to remove abnormal data in the data; using the data storage module to transmit the pre-processed real-time energy consumption data to a database for persistent storage to obtain historical energy consumption data; using the energy consumption analysis module to analyze the historical energy consumption data in the database to generate an energy consumption report; using the energy consumption prediction module to predict the data of a predetermined stage in the future based on the historical energy consumption data to obtain an energy consumption prediction result; using the energy consumption alarm module to issue an alarm signal when the energy consumption prediction result exceeds a threshold; using the user management module to manage system user permissions to ensure system security and access control.
[0017] In one embodiment, the energy consumption prediction module is used to predict the data of a future scheduled stage based on historical energy consumption data to obtain an energy consumption prediction result, which specifically includes: taking the time length of the future scheduled stage as a window, dividing the historical energy consumption data to obtain a number of window energy consumption data arranged in chronological order; inputting the last window energy consumption data into a pre-trained CNN-LSTM model for prediction to obtain an energy consumption prediction result corresponding to the future scheduled stage.
[0018] In one embodiment, the future scheduled period is specifically: the next day, or the next week, or the next month, or the next quarter.
[0019] In one embodiment, the window energy consumption data is specifically a two-dimensional data matrix; the window energy consumption data specifically includes: energy consumption statistical data of each sensor at each sampling time point.
[0020] In one embodiment, the energy consumption data of the last window is input into a pre-trained CNN-LSTM model for prediction to obtain an energy consumption prediction result corresponding to a future scheduled stage, specifically including: inputting the window energy consumption data into a CNN sub-model for convolution to obtain a corresponding energy consumption feature map; inputting the energy consumption feature map into a pooling layer for pooling, and inputting the pooled feature map into a flattening layer for flattening to obtain a one-dimensional vector; inputting the one-dimensional vector into an LSTM sub-model to analyze the time dependence of the one-dimensional vector, and generating an energy consumption prediction result corresponding to a future scheduled stage.
[0021] In one embodiment, the one-dimensional vector is input into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generate an energy consumption prediction result corresponding to a predetermined future stage, specifically including: adjusting the data shape of the one-dimensional vector according to the input rule of the first LSTM layer of the LSTM sub-model to obtain an energy consumption feature vector; inputting the energy consumption feature vector into the first LSTM layer to learn long-term dependencies, and returning the first hidden state sequence of each time step in the first LSTM layer; inputting the first hidden state sequence into the Dropout layer, and randomly discarding a predetermined number of data in the first hidden state sequence; inputting the first hidden state sequence after discarding the data into the second LSTM layer to extract time dependency features, and returning the energy consumption prediction feature vector corresponding to the hidden state of the last time step in the second LSTM layer.
[0022] In one embodiment, the one-dimensional vector is input into the LSTM sub-model to analyze the time dependence of the one-dimensional vector to generate an energy consumption prediction result corresponding to a future predetermined stage, and also includes: inputting the energy consumption prediction feature vector into a fully connected layer to generate a final energy consumption prediction result; the energy consumption prediction result is specifically the energy consumption detection data corresponding to each energy consumption sensor in the future predetermined stage.
[0023] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.
[0024] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.
[0025] In summary, the present invention has the following beneficial effects: a digital factory energy consumption management system, comprising: a data acquisition module; a data transmission module; a data processing module; a data storage module; an energy consumption analysis module; an energy consumption prediction module; an energy consumption alarm module; and a user management module; the energy consumption management system of the present invention can realize automatic recording of factory energy consumption, and can also accurately predict the factory's energy consumption in the future stage. If the factory's energy consumption exceeds a predetermined range, an alarm will be issued in time to avoid the factory's production work being affected by excessive short-term energy consumption demand. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 A structural diagram of a digital factory energy consumption management system of the present invention;
[0027] Figure 2 A flow chart of a digital factory energy consumption management method in an embodiment of the present invention;
[0028] Figure 3 1 is a diagram showing the internal structure of a computer device in an embodiment of the present invention.
[0029] In the figure: 1. Data acquisition module; 2. Data transmission module; 3. Data processing module; 4. Data storage module; 5. Energy consumption analysis module; 6. Energy consumption prediction module; 7. Energy consumption alarm module; 8. User management module. DETAILED DESCRIPTION
[0030] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. Several embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein.
[0031] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.
[0032] In the present invention, unless otherwise clearly specified and limited, the first feature being "above" or "below" the second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Moreover, the first feature being "above", "above" and "above" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature includes the first feature being directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature. The terms "vertical", "horizontal", "left", "right", "above", "below" and similar expressions are for illustrative purposes only, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.
[0033] Embodiment 1
[0034] In order to solve the above problems, the present invention provides a digital factory energy consumption management system, such as Figure 1 As shown, including:
[0035] Data collection module 1: used to collect real-time energy consumption data of equipment in each area; including but not limited to electricity consumption, gas consumption, water resource use, etc. In this embodiment, the most important thing is to collect and detect the power consumption of the equipment. The detection of power consumption can be done by setting up a separate power statistics table for each device, or directly using an electric meter to detect the input end of the device. In the event of network anomalies or failures, the data collection module can also temporarily store data and upload it after the network is restored to prevent data loss. Real-time monitoring of the energy consumption of the equipment provides timely and accurate data for subsequent data analysis and prediction.
[0036] Data transmission module 2: used to upload the collected real-time energy consumption data to the central system to ensure reliable data transmission and centralized storage; in order to meet different environments, the data transmission module can use a wireless communication module for data transmission, or it can use wired communication for stability. Among them, wireless transmission can use Wi-Fi, LoRa, cellular mobile network, Zigbee and other communication protocols. In order to improve the security of data transmission and avoid data leakage, it is also necessary to encrypt the data during the transmission process to ensure the confidentiality and security of the data. Ensure that the energy consumption data can be transmitted to the central system in a timely and accurate manner, providing guarantee for subsequent data processing and analysis.
[0037] Data processing module 3: used to pre-process the real-time energy consumption data received by the central system, including data format conversion, data cleaning, outlier processing, etc., using algorithms (such as statistical methods, machine learning algorithms) to identify and filter abnormal data, such as mutation values and data noise caused by sensor failures. For missing or damaged data, interpolation, prediction and other methods are used to complete the data to ensure data continuity. Data processing module 3 improves data quality and provides a reliable data basis for subsequent data analysis, prediction and decision-making. Eliminate the impact of errors and abnormal data on system analysis results to ensure the accuracy and reliability of energy consumption management.
[0038] Data storage module 4: used to transfer the pre-processed real-time energy consumption data to the database for persistent storage to obtain historical energy consumption data; archive and classify the historical data for long-term storage and query; provide historical data support for energy consumption analysis and prediction, and preserve the company's important energy consumption data assets; data storage module 4 can also provide efficient data management and retrieval functions, supporting fast data query and analysis.
[0039] Energy consumption analysis module 5: used to analyze the historical energy consumption data in the database and generate energy consumption reports; supports energy consumption analysis by time (hours, days, weeks, months, years), regions, equipment, production lines and other dimensions, and intuitively displays energy consumption data and analysis results through charts (such as line charts, bar charts, pie charts, heat maps, etc.) to facilitate user understanding and decision-making, help enterprises fully understand the energy consumption status, identify high-energy consumption equipment and links, provide decision-making basis for management, and formulate energy-saving and emission reduction strategies and optimization measures.
[0040] Energy consumption prediction module 6: used to predict the data of the future scheduled stage based on the historical energy consumption data to obtain the energy consumption prediction results; based on the historical energy consumption data, use machine learning or deep learning algorithms (such as CNN-LSTM model) to establish an energy consumption prediction model, predict the energy consumption of the future scheduled stage (such as the next day, next week, next month), provide accurate energy consumption estimates, support energy consumption simulation under different production plans, seasons, equipment operation strategies and other conditions, help enterprises with energy planning, and enterprises can adjust production arrangements to avoid energy consumption peaks and reduce energy costs.
[0041] Energy consumption alarm module 7: used to send out an alarm signal when the energy consumption prediction result exceeds the threshold; allows users to set energy consumption thresholds according to needs, including real-time energy consumption thresholds, cumulative energy consumption thresholds, predicted energy consumption thresholds, etc., continuously monitors actual energy consumption and predicted energy consumption, and compares them with the set thresholds. When the energy consumption data exceeds the set threshold or an abnormal situation occurs, the alarm mechanism is automatically triggered, and relevant personnel are notified through various methods, such as system pop-ups, text messages, emails, mobile application push, etc., to ensure timely response; timely detection and response to abnormal energy consumption, to prevent energy waste and potential equipment failures, to improve the initiative of energy consumption management, and to prevent exceeding the energy budget or violating energy regulations.
[0042] User Management Module 8: Used to manage system user permissions and ensure system security and access control. Use username and password, two-factor authentication, etc. to ensure that only authorized users can access the system; assign different access rights and operation rights according to the user's role and responsibilities to ensure the security of data and functions; define and manage different user roles (such as system administrators, energy managers, production managers, etc.) to facilitate permission allocation and management; record user login, operation, modification and other behaviors, provide audit and traceability functions, and improve system security and compliance; support password policies (such as password complexity, validity period), account lock policies, permission change review, etc. to enhance system security protection.
[0043] Specifically, in an injection molding plant, various equipment consumes a lot of energy. For example, the heating system, hydraulic system, and motor drive system of the injection molding machine used to produce plastic accessories consume a lot of electrical energy. Drying equipment, including hopper dryers, dehumidifying dryers, etc., are used to remove moisture from plastic raw materials. They usually rely on electric heating, resulting in a large amount of electrical energy consumption. Cooling equipment: including industrial chillers, cooling towers, etc., are used to cool equipment and products. In contrast to heating equipment, cooling equipment also consumes a lot of energy to adjust the temperature. Air compression equipment, used to power the pneumatic components of the equipment, needs to run for a long time to maintain pressure, and therefore consumes a lot of energy. There are other equipment, such as conveyor belts, mixers, crushers, lighting and air conditioning in the factory, which are also important sources of energy consumption. Although the energy consumption is not high, the number is large and the cumulative energy consumption is significant.
[0044] In summary, this application provides a complete digital factory energy consumption management system, covering the entire process of energy consumption data collection, transmission, processing, storage, analysis, prediction, alarm and user management. Through real-time monitoring and analysis, energy saving potential can be discovered, energy use can be optimized, and operating costs can be reduced; accurate data and predictions can be provided to assist management in formulating energy strategies and production plans, and abnormal energy consumption alarms can be used to prevent equipment failures and safety accidents. Strengthen system security: Through strict user management, corporate data and system security can be protected to prevent data leakage and abuse.
[0045] Embodiment 2
[0046] See also Figure 2 , a digital factory energy consumption management method, applied to the digital factory energy consumption management system according to claim 1, the management method specifically includes:
[0047] S1. Using the data acquisition module to collect real-time energy consumption data of equipment in each area;
[0048] S2. Uploading the collected real-time energy consumption data to the central system using the data transmission module to ensure reliable transmission and centralized storage of data;
[0049] S3, using the data processing module to pre-process the real-time energy consumption data received by the central system to remove abnormal data in the data;
[0050] S4, using the data storage module to transfer the preprocessed real-time energy consumption data to a database for persistent storage to obtain historical energy consumption data;
[0051] S5. Analyze the historical energy consumption data in the database using the energy consumption analysis module to generate an energy consumption report;
[0052] S6, using the energy consumption prediction module to predict the data of the future predetermined stage according to the historical energy consumption data, to obtain the energy consumption prediction result;
[0053] S7, using the energy consumption alarm module to issue an alarm signal when the energy consumption prediction result exceeds a threshold;
[0054] S8. Use the user management module to manage system user rights to ensure system security and access control.
[0055] Specifically, in this embodiment, a digital factory energy consumption management system provided in Embodiment 1 is used to further provide a digital factory energy consumption management method. Using this method, real-time and accurate collection and monitoring of energy consumption data can be achieved, and the energy usage of the factory can be fully understood; advanced algorithms and models are used to conduct in-depth analysis and accurate prediction of energy consumption data to help enterprises optimize energy management. According to the prediction results, energy consumption anomalies can be discovered and warned in a timely manner to prevent risks and ensure the safety and continuity of production; through strict user authority management and modular design, the security, stability and scalability of the system are ensured, energy consumption and operating costs are effectively reduced, and the economic and environmental benefits of the enterprise are improved.
[0056] In one embodiment, the energy consumption prediction module is used to predict data of a future predetermined stage based on historical energy consumption data to obtain energy consumption prediction results, specifically including:
[0057] Taking the time length of the future scheduled stage as a window, segmenting the historical energy consumption data to obtain a number of window energy consumption data arranged in chronological order;
[0058] The energy consumption data of the last window is input into the pre-trained CNN-LSTM model for prediction, and the energy consumption prediction results corresponding to the future scheduled stage are obtained.
[0059] Specifically, in this embodiment, by using a time window of the same length as the future prediction stage, historical data is segmented so that the model can better learn patterns and features related to the prediction target and improve prediction accuracy. Window data are arranged in chronological order to ensure the continuity of time dependencies, which can enhance the model's ability to capture time features. In the prediction process, the data of the last window is used as input data, which can make full use of recent energy consumption trends and change information and enhance the reliability of short-term predictions.
[0060] In this embodiment, the CNN-LSTM model is used to predict energy consumption data, and the convolutional neural network (CNN) is good at extracting local features from data. In time series data, CNN can identify short-term patterns, trends, seasonal and periodic changes, where the convolution kernel of CNN can adjust the corresponding size according to the time window. When the time window is relatively small, for one day, the size of the corresponding CNN convolution kernel is also relatively small; when the time window is relatively large, for one day, the size of the corresponding CNN convolution kernel is also relatively small. Through the convolution operation, CNN can convert the original data into a higher-level feature representation, extract key information, and reduce noise interference. After convolution and pooling, the time step input to LSTM can be reduced, reducing the computational complexity of LSTM, and the features extracted by CNN are more refined than the original energy consumption data, making it easier for LSTM to learn important patterns in time series.
[0061] In one embodiment, the future scheduled period is specifically: the next day, or the next week, or the next month, or the next quarter.
[0062] Specifically, this application supports multiple forecasting time scales to meet the needs of enterprises at different planning and decision-making stages, such as short-term scheduling and long-term energy planning. According to different time spans, enterprises can formulate more sophisticated energy consumption control and optimization strategies.
[0063] In one embodiment, the window energy consumption data is specifically a two-dimensional data matrix; the window energy consumption data specifically includes: energy consumption statistical data of each sensor at each sampling time point.
[0064] By using a two-dimensional matrix to fuse data from the time and sensor space dimensions, the model can simultaneously perceive the energy consumption characteristics in time and space, and follow the changing trends over time through data from different sensors, which helps the model to more accurately predict overall and local energy consumption trends.
[0065] In one embodiment, the energy consumption data of the last window is input into a pre-trained CNN-LSTM model for prediction to obtain an energy consumption prediction result corresponding to a future scheduled stage, specifically including: inputting the window energy consumption data into a CNN sub-model for convolution to obtain a corresponding energy consumption feature map; inputting the energy consumption feature map into a pooling layer for pooling, and inputting the pooled feature map into a flattening layer for flattening to obtain a one-dimensional vector; inputting the one-dimensional vector into an LSTM sub-model to analyze the time dependence of the one-dimensional vector, and generating an energy consumption prediction result corresponding to a future scheduled stage.
[0066] In specific use, this application uses the convolution operation of the CNN sub-model to extract local patterns in the energy consumption data, such as periodic changes and local anomalies, to improve the feature representation capability; the energy consumption feature map after feature extraction is input into the pooling layer to reduce the data dimension, retain important features, and reduce the model complexity; the LSTM sub-model performs time series analysis on the flattened one-dimensional vector to capture long-term and short-term time dependencies and enhance the accuracy of the prediction. Combining the spatial feature extraction of CNN and the time series modeling capabilities of LSTM, it can give full play to the advantages of both and improve the overall prediction performance.
[0067] In one embodiment, the one-dimensional vector is input into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generate an energy consumption prediction result corresponding to a predetermined future stage, specifically including: adjusting the data shape of the one-dimensional vector according to the input rule of the first LSTM layer of the LSTM sub-model to obtain an energy consumption feature vector; inputting the energy consumption feature vector into the first LSTM layer to learn long-term dependencies, and returning the first hidden state sequence of each time step in the first LSTM layer; inputting the first hidden state sequence into the Dropout layer, and randomly discarding a predetermined number of data in the first hidden state sequence; inputting the first hidden state sequence after discarding the data into the second LSTM layer to extract time dependency features, and returning the energy consumption prediction feature vector corresponding to the hidden state of the last time step in the second LSTM layer.
[0068] Specifically, by adjusting the data shape to ensure that the input meets the requirements of the LSTM layer and avoiding data format errors, the stability of the model is improved. A double-layer LSTM is used, in which the first LSTM layer learns the basic long-term dependencies, and the second LSTM layer further extracts high-level temporal features to capture complex time series patterns. A Dropout layer is introduced between the first LSTM layer and the second LSTM layer to randomly discard part of the hidden state data, reduce the model's excessive dependence on training data, and improve the model's generalization ability. The hidden state of the last time step of the second LSTM layer is used to focus on the most important recent information in the time series, thereby enhancing the real-time and accuracy of the prediction.
[0069] The high-dimensional convolutional features are converted into one-dimensional vectors to facilitate processing in the subsequent fully connected layer, but the flattening operation will destroy the structural information of the time series, which may affect the model's capture of time dependencies in the subsequent LSTM layer. Therefore, before inputting into the first LSTM layer, the flattened one-dimensional vector needs to be resized to the dimension required by the first LSTM layer.
[0070] Specifically, the hidden units in the first LSTM layer receive the adjusted energy consumption feature vectors and update the state through the forget gate, input gate, output gate and memory unit in the first LSTM layer. In the case of the current time step, the LSTM layer receives the input x of the current time step. t , the hidden state h of the previous time step t-1 and the memory cell state C t-1 ; Multiple hidden units will process the same input information in parallel. The time step refers to the energy consumption data corresponding to the sampling point of each time dimension input into the model;
[0071] Since multiple hidden units are independent of each other, the data processing process mainly includes:
[0072] The forget gate activation value of the jth hidden unit in the tth time step Specifically:
[0073]
[0074] Among them, σ represents the Sigmoid activation function; is the forget gate weight matrix of the jth hidden unit; is the bias term, the activation value of the forget gate Used to control the information that needs to be forgotten in the state of the memory unit.
[0075] The input gate activation value of the jth hidden unit at the tth time step Specifically:
[0076]
[0077] Among them, σ represents the Sigmoid activation function; is the input gate weight matrix of the jth hidden unit; is the bias term, the input gate activation value Used to control which new information in the current time step t needs to be written into the memory cell.
[0078] In the tth time step, the jth hidden unit calculates its candidate memory unit state
[0079]
[0080] Use the candidate memory state to update the memory unit state of the jth hidden unit at the previous time step t-1 Get the memory cell state at the current time step:
[0081]
[0082] The memory cell state means that each hidden unit can independently decide what information to keep or forget in its memory cell state.
[0083] The output gate activation value of the jth hidden unit at the tth time step Specifically:
[0084]
[0085] Among them, σ represents the Sigmoid activation function; is the output gate weight matrix of the jth hidden unit; is the bias term, the output gate activation value Used to control the information output from the memory unit.
[0086] In the tth time step, the jth hidden unit calculates the hidden state of the current time step t, including:
[0087]
[0088] The adjusted energy consumption feature vector is outputted by all hidden units at each time step. t , combined into an output vector, that is, the first hidden state sequence; in order to avoid overfitting of the model, the first hidden state sequence needs to be input into the Dropout layer to randomly discard a predetermined amount of data, which is 30% in this application, to improve the generalization ability of the model. The discarding process keeps the shape of the vector unchanged, but some data becomes 0. The first hidden state sequence after the discarded data is then input into the second LSTM layer for processing, and finally outputs the energy consumption data sequence information output at the last time step. The purpose is to summarize the information of the entire time series, and predict the energy consumption usage within a certain period of time in the future, and provide input to the subsequent fully connected layer so that the subsequent fully connected layer can output a readable energy consumption prediction result.
[0089] In one embodiment, the one-dimensional vector is input into the LSTM sub-model to analyze the time dependence of the one-dimensional vector to generate an energy consumption prediction result corresponding to a future predetermined stage, and also includes: inputting the energy consumption prediction feature vector into a fully connected layer to generate a final energy consumption prediction result; the energy consumption prediction result is specifically the energy consumption detection data corresponding to each energy consumption sensor in the future predetermined stage.
[0090] In actual use, the fully connected layer maps high-dimensional feature vectors to specific prediction results, realizes the comprehensive and nonlinear combination of features, and improves the prediction accuracy. In addition, the fully connected layer can directly output the prediction value of each energy consumption sensor according to actual needs, and provide fine-grained energy consumption prediction information, which is convenient for formulating energy-saving measures for different equipment. The generated prediction results are intuitive and clear, which is convenient for managers to understand and apply, and improves the readability of the system.
[0091] Embodiment 3
[0092] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, a digital factory energy consumption management method as described in the second embodiment is implemented.
[0093] Embodiment 4
[0094] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the computer program is executed by the processor, a digital factory energy consumption management method is implemented.
[0095] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0096] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: The management method specifically includes:
[0097] S1. Using the data acquisition module to collect real-time energy consumption data of equipment in each area;
[0098] S2. Uploading the collected real-time energy consumption data to the central system using the data transmission module to ensure reliable transmission and centralized storage of data;
[0099] S3, using the data processing module to pre-process the real-time energy consumption data received by the central system to remove abnormal data in the data;
[0100] S4, using the data storage module to transfer the preprocessed real-time energy consumption data to a database for persistent storage to obtain historical energy consumption data;
[0101] S5. Analyze the historical energy consumption data in the database using the energy consumption analysis module to generate an energy consumption report;
[0102] S6, using the energy consumption prediction module to predict the data of the future predetermined stage according to the historical energy consumption data, to obtain the energy consumption prediction result;
[0103] S7, using the energy consumption alarm module to issue an alarm signal when the energy consumption prediction result exceeds a threshold;
[0104] S8. Use the user management module to manage system user rights to ensure system security and access control.
[0105] In one embodiment, the energy consumption prediction module is used to predict data of a future predetermined stage based on historical energy consumption data to obtain energy consumption prediction results, specifically including:
[0106] Taking the time length of the future scheduled stage as a window, segmenting the historical energy consumption data to obtain a number of window energy consumption data arranged in chronological order;
[0107] The energy consumption data of the last window is input into the pre-trained CNN-LSTM model for prediction, and the energy consumption prediction results corresponding to the future scheduled stage are obtained.
[0108] In one embodiment, the future scheduled period is specifically: the next day, or the next week, or the next month, or the next quarter.
[0109] In one embodiment, the window energy consumption data is specifically a two-dimensional data matrix; the window energy consumption data specifically includes: energy consumption statistical data of each sensor at each sampling time point.
[0110] In one embodiment, the energy consumption data of the last window is input into a pre-trained CNN-LSTM model for prediction to obtain the energy consumption prediction result corresponding to the future predetermined stage, specifically including:
[0111] Input the window energy consumption data into the CNN sub-model for convolution to obtain a corresponding energy consumption feature map;
[0112] Inputting the energy consumption feature map into the pooling layer for pooling, and inputting the pooled feature map into the flattening layer for flattening to obtain a one-dimensional vector;
[0113] The one-dimensional vector is input into the LSTM sub-model to analyze the time dependency of the one-dimensional vector, and generate energy consumption prediction results corresponding to the future predetermined stage.
[0114] In one embodiment, the inputting of the one-dimensional vector into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generate the energy consumption prediction result corresponding to the future predetermined stage specifically includes:
[0115] According to the input rule of the first LSTM layer of the LSTM sub-model, the data shape of the one-dimensional vector is adjusted to obtain an energy consumption feature vector;
[0116] Input the energy consumption feature vector into the first LSTM layer to learn the long-term dependency, and return the first hidden state sequence of each time step in the first LSTM layer;
[0117] Inputting the first hidden state sequence into a Dropout layer, and randomly discarding a predetermined amount of data in the first hidden state sequence;
[0118] The first hidden state sequence after discarding the data is input into the second LSTM layer to extract the time-dependent features, and the energy consumption prediction feature vector corresponding to the hidden state of the last time step in the second LSTM layer is returned.
[0119] In one embodiment, the step of inputting the one-dimensional vector into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generating an energy consumption prediction result corresponding to a future predetermined stage further includes:
[0120] The energy consumption prediction feature vector is input into the fully connected layer to generate a final energy consumption prediction result; the energy consumption prediction result is specifically the energy consumption detection data corresponding to each energy consumption sensor in a predetermined future stage.
[0121] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0123] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.
Claims
1. A digital factory energy consumption management system, characterized in that: include: Data collection module: used to collect real-time energy consumption data of equipment in each area; Data transmission module: used to upload the collected real-time energy consumption data to the central system to ensure reliable data transmission and centralized storage; Data processing module: used to pre-process the real-time energy consumption data received by the central system and remove abnormal data; Data storage module: used to transfer the pre-processed real-time energy consumption data to the database for persistent storage to obtain historical energy consumption data; Energy consumption analysis module: used to analyze the historical energy consumption data in the database and generate energy consumption reports; Energy consumption prediction module: used to predict the data of the future scheduled stage based on the historical energy consumption data to obtain the energy consumption prediction results; Energy consumption alarm module: used for sending an alarm signal when the energy consumption prediction result exceeds the threshold; User management module: used to manage system user permissions and ensure system security and access control.
2. A digital factory energy consumption management method, applied to the digital factory energy consumption management system according to claim 1, characterized in that: The management method specifically includes: Using the data acquisition module to collect real-time energy consumption data of equipment in each area; The data transmission module is used to upload the collected real-time energy consumption data to the central system to ensure reliable transmission and centralized storage of data; The data processing module is used to pre-process the real-time energy consumption data received by the central system to remove abnormal data in the data; The data storage module is used to transfer the preprocessed real-time energy consumption data to the database for persistent storage to obtain historical energy consumption data; Analyze the historical energy consumption data in the database using the energy consumption analysis module to generate an energy consumption report; The energy consumption prediction module is used to predict the data of the future predetermined stage according to the historical energy consumption data to obtain the energy consumption prediction result; Using the energy consumption alarm module to send out an alarm signal when the energy consumption prediction result exceeds a threshold value; The user management module is utilized to manage system user rights to ensure system security and access control.
3. A digital factory energy consumption management method according to claim 2, characterized in that: The energy consumption prediction module is used to predict the data of a future predetermined stage according to the historical energy consumption data to obtain the energy consumption prediction result, specifically including: Taking the time length of the future scheduled stage as a window, segmenting the historical energy consumption data to obtain a number of window energy consumption data arranged in chronological order; The energy consumption data of the last window is input into the pre-trained CNN-LSTM model for prediction, and the energy consumption prediction results corresponding to the future scheduled stage are obtained.
4. A digital factory energy consumption management method according to claim 3, characterized in that: The future scheduled stage is specifically: the next day, or the next week, or the next month, or the next quarter.
5. A digital factory energy consumption management method according to claim 4, characterized in that: The window energy consumption data is specifically a two-dimensional data matrix; The window energy consumption data specifically includes: energy consumption statistics of each sensor at each sampling time point.
6. A digital factory energy consumption management method according to claim 5, characterized in that: The energy consumption data of the last window is input into the pre-trained CNN-LSTM model for prediction, and the energy consumption prediction result corresponding to the future predetermined stage is obtained, which specifically includes: Input the window energy consumption data into the CNN sub-model for convolution to obtain a corresponding energy consumption feature map; Inputting the energy consumption feature map into the pooling layer for pooling, and inputting the pooled feature map into the flattening layer for flattening to obtain a one-dimensional vector; The one-dimensional vector is input into the LSTM sub-model to analyze the time dependency of the one-dimensional vector, and generate energy consumption prediction results corresponding to the future predetermined stage.
7. A digital factory energy consumption management method according to claim 6, characterized in that: The inputting of the one-dimensional vector into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generate the energy consumption prediction result corresponding to the future predetermined stage specifically includes: According to the input rule of the first LSTM layer of the LSTM sub-model, the data shape of the one-dimensional vector is adjusted to obtain an energy consumption feature vector; Input the energy consumption feature vector into the first LSTM layer to learn the long-term dependency, and return the first hidden state sequence of each time step in the first LSTM layer; Inputting the first hidden state sequence into a Dropout layer, and randomly discarding a predetermined amount of data in the first hidden state sequence; The first hidden state sequence after discarding the data is input into the second LSTM layer to extract the time-dependent features, and the energy consumption prediction feature vector corresponding to the hidden state of the last time step in the second LSTM layer is returned.
8. A digital factory energy consumption management method according to claim 7, characterized in that: The step of inputting the one-dimensional vector into the LSTM sub-model to analyze the time dependency of the one-dimensional vector and generate an energy consumption prediction result corresponding to a future predetermined stage also includes: The energy consumption prediction feature vector is input into the fully connected layer to generate a final energy consumption prediction result; the energy consumption prediction result is specifically the energy consumption detection data corresponding to each energy consumption sensor in a predetermined future stage.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the digital factory energy consumption management method as described in any one of claims 2 to 8 is implemented.
10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the digital factory energy consumption management method as described in any one of claims 2 to 8 is implemented.