Modeling analysis system based on digital energy configuration

Abstract

The invention discloses a modeling analysis system based on digital energy configuration, and relates to the technical field of energy consumption management analysis, and the system comprises an energy consumption model which is established based on production data of energy consumption feature point grouping, equipment operation data and equipment operation environment data, and reflects the changes of the production data and the equipment operation data through the energy consumption model, the influence of the equipment operation environment on the energy consumption model is determined, the equipment state is predicted by combining the production data and the change of the equipment operation data, and production optimization suggestions are made; the decision-making model is used for carrying out optimization decision-making according to analysis of the energy consumption model on the energy consumption points, determining decision-making cost according to the decision-making and optimizing calculation of flow energy consumption and efficiency; according to the method, the energy consumption of the production line is predicted, the optimization suggestion for the production line can be made according to the energy consumption of the production line, the energy consumption hot spots are determined according to the energy consumption data corresponding to different equipment operation data and production data, and the process steps are optimized according to the energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption management and analysis, and in particular to a modeling and analysis system based on digital energy configuration. Background Art

[0002] In the process of enterprise development, traditional enterprise management is mainly decided by people, implemented and executed. Its efficiency is limited by many factors, which seriously affects the development speed of the enterprise and makes the judgment of enterprise development more subjective. Due to the intervention of scientific and technological products such as computers, work efficiency has been greatly increased, thus leading to technological changes in other aspects of the enterprise. Artificial intelligence came into being under such circumstances. It conducts in-depth analysis based on factory operation data and provides suggestions for energy-saving transformation or cost reduction and efficiency improvement.

[0003] The relationship between production line energy consumption and production speed, temperature, and equipment status is of great significance for optimizing the production process, improving efficiency, and reducing costs.

[0004] Announcement No. CN205231853U discloses an Internet of Things-based power consumption analysis system, which consists of a power consumption analyzer and a power consumption monitor. The power consumption analyzer includes a wireless communication module, an energy consumption analyzer, an energy-saving potential analyzer, and an economic analyzer. The power consumption monitor includes an energy-consuming equipment monitor, a power quality monitor, a wireless communication module, a microprocessor, and a power supply. The power consumption analysis system based on the Internet of Things has the advantages of simple structure, intelligence, humanization, timeliness and effectiveness, etc. It can monitor energy-consuming equipment and power quality in real time, and can perform energy consumption analysis, energy-saving potential analysis, and economic analysis, and can accurately predict and guide energy consumption for enterprises.

[0005] In the existing power consumption analysis system, energy consumption is analyzed by monitoring power-consuming equipment and power quality, such as the power consumption analysis system based on the Internet of Things disclosed above. However, for a production line, there are many parameters related to energy consumption, such as production speed, temperature, equipment status and other data. Monitoring only by monitoring power-consuming equipment and power quality cannot determine the energy consumption points in the production line and process, and provide intelligent suggestions for production management personnel. Therefore, a modeling and analysis system based on digital energy configuration is proposed, which is used to determine the energy consumption points in the production line and provide intelligent suggestions based on the energy consumption points. Summary of the invention

[0006] One of the objectives of the present invention is to provide a modeling and analysis system based on digital energy configuration.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A modeling and analysis system based on digital energy configuration, comprising:

[0008] Data acquisition module, which obtains ERP data and MES data, and determines production data, energy consumption data, and equipment operation status data;

[0009] The data cleaning module cleans the production data, energy consumption data and equipment operation status data collected by the data acquisition module, removes redundant data and determines data integrity, repairs missing values ​​according to the changes in the data before and after the missing data, and adjusts the data format standardization;

[0010] The data integration module integrates different data, establishes a data source association network, and associates production data, energy consumption data, and equipment operation data;

[0011] The data management module stores and backs up related data according to the data source association network and sets data access permissions to protect different types of data;

[0012] The data preprocessing module determines the energy consumption characteristic points and groups the production data and equipment operation status data according to the energy consumption characteristic points;

[0013] Energy consumption model: Establish energy consumption model based on production data, equipment operation data and equipment operation environment data grouped by energy consumption feature points. The energy consumption model reflects changes in production data and equipment operation data, and determines the impact of the equipment operation environment on the energy consumption model. Combined with changes in production data and equipment operation data, the equipment status is predicted and production optimization suggestions are made.

[0014] Decision-making model, which makes optimization decisions based on the analysis of energy consumption points by the energy consumption model, determines the decision cost based on the decision, and calculates the energy consumption and efficiency of the optimization process;

[0015] The execution unit controls external devices according to the optimization decision and obtains the data changes after the adjustment decision to feedback the execution result.

[0016] In one or more embodiments of the present invention, ERP data and MES data include set data and real-time data, the set data being the production plan, and the real-time data being the equipment data changes obtained through sensors installed on the equipment, as well as the energy consumption data of the equipment at different times.

[0017] In one or more embodiments of the present invention, the data cleaning module performs the following steps for data cleaning:

[0018] Compare key fields in production data, energy consumption data, and equipment operation data, identify duplicate data, and delete identical or redundant data records based on the identification results;

[0019] Detect missing values, scan the data set, determine the location and number of missing values, and use interpolation algorithms to infer missing values ​​based on previous and subsequent data;

[0020] Detect outliers in the data set through statistical analysis and correct them according to reasonable ranges or trends;

[0021] Adjust the data format of the dataset so that the naming and format of the data fields are consistent.

[0022] In one or more embodiments of the present invention, the steps of establishing the data source association network are as follows:

[0023] Determine the data source and clarify all data sources that need to be associated, including: production data, equipment operation data and energy consumption data, and determine the abnormal energy consumption points in the energy consumption data;

[0024] Select appropriate associated fields, match energy consumption data with production data and equipment operation data through associated fields, and establish a data source association network;

[0025] The data source association network is marked with a timestamp and stored according to the order of the timestamps.

[0026] In one or more embodiments of the present invention, key variables that can significantly affect energy consumption are identified and defined: production speed, equipment operating temperature, and equipment operating status. By analyzing historical energy consumption data, energy consumption characteristic points during high energy consumption periods and under specific operating conditions are identified, and the production data and equipment operating data corresponding to the energy consumption characteristic points are obtained, and the production data and equipment operating data corresponding to the energy consumption characteristic points are grouped.

[0027] In one or more embodiments of the present invention, the energy consumption model establishment method is as follows:

[0028] Obtain each set of data corresponding to the energy consumption characteristic point, build a random forest model for each set of data, and predict the energy consumption under the production conditions or equipment status;

[0029] Use the data set to optimize each model and optimize model parameters to improve prediction accuracy;

[0030] Cross-validation evaluates the performance of each model to avoid overfitting;

[0031] Combine multiple random forest models into an energy consumption model, and combine the prediction results of each model by weighted average or stacking;

[0032] According to the performance of each random forest model validation set, the weight of each random forest model is assigned to ensure that the prediction accuracy of the energy consumption model is maximized.

[0033] In one or more embodiments of the present invention, a validation set is used to evaluate the performance of each random forest model, and a weight is calculated based on the performance index of the random forest model:

[0034] Calculate the error of each model relative to the validation machine;

[0035] Calculate the inverse of the error as the preliminary weight, and normalize the preliminary weight so that its sum is 1;

[0036] The random forest model weights are calculated as follows:

[0037]

[0038] where w i is the weight of the i-th random forest model, MSE i is the mean square error of the ith random forest model, is the sum of the inverse errors of all random forest models;

[0039] Combine the predictions of each model using all random forest model weights;

[0040] Determine the final predicted value I of the energy consumption model;

[0041] I=w a I a +w b I b +…+w n I n ;

[0042] Among them, I n is the prediction result of the nth random forest model.

[0043] In one or more embodiments of the present invention, the steps of obtaining device operating environment data and real-time device operating data and predicting changes in device operating data through the device operating environment data are as follows:

[0044] Determine the factors that affect the real-time equipment operation data and equipment operation environment data on the equipment operation data;

[0045] Calculate the impact of equipment operating environment data on equipment operating data;

[0046] Calculate the change range and change speed of equipment operation data based on the impact value;

[0047] Obtain the difference between the change of the equipment operation data in the subsequent time and the change amplitude and change speed of the calculated equipment operation data, and correct the equipment operation data of the predicted time according to the difference.

[0048] In one or more embodiments of the present invention, production data and equipment operation data of each process are obtained, and the total energy consumption of each process is obtained, and the energy consumption per unit output is calculated according to the production speed and output, and the energy consumption baseline of each process under normal conditions is established;

[0049] Determine the energy consumption overlap area of ​​the process, and use the energy consumption model to analyze the energy consumption overlap area of ​​the process, split the process steps, combine the process steps with overlapping areas and perform energy consumption simulation;

[0050] Use statistical methods and machine learning algorithms to identify energy consumption anomalies in energy consumption simulations;

[0051] Propose production optimization suggestions based on abnormal energy consumption analysis results.

[0052] In one or more embodiments of the present invention, the decision model performs simulation based on the production optimization suggestion;

[0053] Design multiple optimization plans based on process flow and equipment operating status, including equipment maintenance, production scheduling, and process parameter adjustment;

[0054] Use the decision-making model to simulate each optimization plan and predict the changes in energy consumption, production efficiency, and cost indicators corresponding to each optimization plan;

[0055] Calculate the cost of equipment maintenance and production downtime based on the implementation details of each optimization plan;

[0056] Evaluate the expected benefits of each optimization solution: reduced energy consumption, increased production efficiency, and improved product quality;

[0057] Calculate the consumption and benefits of each optimization plan and determine the plan with the best comprehensive benefits.

[0058] Through the above technical solution, the present invention has the following beneficial effects:

[0059] 1. This application can make optimization suggestions for the production line based on the energy consumption of the production line by predicting the energy consumption of the production line. According to the energy consumption data corresponding to the operation data and production data of different equipment, the energy consumption hotspots can be determined, and the process steps can be optimized according to the energy consumption.

[0060] 2. By integrating data from different sources into a unified format, data collaboration across departments or systems can be carried out, and historical data can be stored and managed to facilitate traceability and long-term analysis. In combination with different levels of user access rights, data security can be protected.

[0061] 3. Through energy consumption models and decision-making models, different production optimization suggestions are simulated, and the advantages and disadvantages of each plan in terms of cost, energy consumption and efficiency are compared, so as to provide intelligent suggestions for production managers, such as adjusting the production line and changing the process, so as to achieve the goal of reducing costs and increasing efficiency.

[0062] 4. Collect feedback data after implementing the recommendations to adjust and improve the analysis model. Continuously optimize the model based on the feedback execution results to adapt to changes in the production environment and new data patterns. Monitor the operating data in real time. When anomalies or deviations from energy-saving targets are found, adjustments can be made through the execution unit.

[0063] Other features and advantages of the present invention will be described in the following description, and part of them will become obvious from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0065] The following will disclose multiple embodiments of the present invention with the accompanying drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. In other words, in some embodiments of the present invention, these practical details are not necessary. And if it is possible to implement, the features of different embodiments can be applied interchangeably.

[0066] Unless otherwise defined, all words (including technical and scientific terms) used herein have their usual meanings, which can be understood by those familiar with this field. Furthermore, the definitions of the above words in commonly used dictionaries should be interpreted as the meanings consistent with the relevant fields of the present invention in the content of this specification. Unless otherwise clearly defined, these words will not be interpreted as idealized or overly formal meanings.

[0067] The above explanation does not completely include the relationship definitions given in this application, but only represents part of it.

[0068] See also Figure 1 The present invention provides a modeling and analysis system based on digital energy configuration, which collects real-time data of factory equipment and environment, establishes an energy consumption model to analyze energy consumption points, and provides an intelligent optimization solution.

[0069] In one embodiment, the modeling and analysis system includes:

[0070] Data acquisition module, which obtains ERP data and MES data, and determines production data, energy consumption data, and equipment operation status data;

[0071] The data cleaning module cleans the production data, energy consumption data and equipment operation status data collected by the data acquisition module, removes redundant data and determines data integrity, repairs missing values ​​according to the changes in the data before and after the missing data, and adjusts the data format standardization;

[0072] The data integration module integrates different data, establishes a data source association network, and associates production data, energy consumption data, and equipment operation data;

[0073] The data management module stores and backs up related data according to the data source association network and sets data access permissions to protect different types of data;

[0074] The data preprocessing module determines the energy consumption characteristic points and groups the production data and equipment operation status data according to the energy consumption characteristic points;

[0075] Energy consumption model: Establish energy consumption model based on production data, equipment operation data and equipment operation environment data grouped by energy consumption feature points. The energy consumption model reflects changes in production data and equipment operation data, and determines the impact of the equipment operation environment on the energy consumption model. Combined with changes in production data and equipment operation data, the equipment status is predicted and production optimization suggestions are made.

[0076] Decision-making model, which makes optimization decisions based on the analysis of energy consumption points by the energy consumption model, determines the decision cost based on the decision, and calculates the energy consumption and efficiency of the optimization process;

[0077] The execution unit controls external devices according to the optimization decision and obtains the data changes after the adjustment decision to feedback the execution result.

[0078] In one feasible method, there is a direct relationship between factory energy consumption and changes in production data, equipment operation data, and equipment operation environment data. By determining the energy consumption point and obtaining the data source association network corresponding to the energy consumption point, it is possible to determine the data change corresponding to the change in the energy consumption point and obtain the change data corresponding to the change in the energy consumption point, thereby being able to analyze the relevant data of abnormal changes in the energy consumption point.

[0079] Among them, the energy consumption model is used to simulate the changes in energy consumption of the production line under different production speeds and different processes. It can simulate the changes in different states and calculate the energy consumption of the production line in different states. The optimization suggestions are determined based on the energy consumption. Different optimization decisions correspond to different cost reductions, efficiency increases and investments. The decision-making model is used to simulate according to the optimization suggestions to determine the energy consumption and efficiency.

[0080] Set user access rights at different levels, and determine the data content that can be accessed according to the different levels to protect data security.

[0081] In one embodiment, ERP data and MES data include set data and real-time data, the set data being the production plan, and the real-time data being the equipment data changes obtained through sensors installed on the equipment, as well as the energy consumption data of the equipment at different times.

[0082] In one feasible method, data collection is performed in two ways to obtain the status of production equipment at different times. The production plan corresponds to the production speed that the equipment needs to achieve per unit time, while the equipment data changes and energy consumption data correspond to the changes in the production speed of the equipment at different times.

[0083] In one embodiment, the data cleaning module performs the following steps for data cleaning:

[0084] Compare key fields in production data, energy consumption data, and equipment operation data, identify duplicate data, and delete identical or redundant data records based on the identification results;

[0085] Detect missing values, scan the data set, determine the location and number of missing values, and use interpolation algorithms to infer missing values ​​based on previous and subsequent data;

[0086] Detect outliers in the data set through statistical analysis and correct them according to reasonable ranges or trends;

[0087] Adjust the data format of the dataset so that the naming and format of the data fields are consistent.

[0088] In one feasible method, the collected data is cleaned to ensure the quality of the data set and to complete the missing values ​​in the data set. When establishing the energy consumption model, the accuracy of the energy consumption model can be guaranteed. By performing data standardization processing, the model can be built faster and more accurately to facilitate energy consumption prediction.

[0089] In one embodiment, the steps of establishing the data source association network are as follows:

[0090] Determine the data source and clarify all data sources that need to be associated, including: production data, equipment operation data and energy consumption data, and determine the abnormal energy consumption points in the energy consumption data;

[0091] Select appropriate associated fields, match energy consumption data with production data and equipment operation data through associated fields, and establish a data source association network;

[0092] The data source association network is marked with a timestamp and stored according to the order of the timestamps.

[0093] Optionally, abnormal energy consumption points in the energy consumption data are determined by defining abnormal values, and the normal range and abnormal threshold of energy consumption are defined based on historical data and industry standards;

[0094] Alternatively, statistical methods are used to identify anomalies in energy consumption data.

[0095] In one feasible method, after the production plan is formulated, the average operating speed of the equipment over a period of time is constant, and the theoretical energy consumption is also constant. The theoretical energy consumption value range is determined, and the energy consumption change curve is obtained. When the energy consumption change curve exceeds the theoretical energy consumption value range, the turning point of the change of the energy consumption change defect is marked as an abnormal energy consumption point.

[0096] In one embodiment, key variables that can significantly affect energy consumption are identified and defined: production speed, equipment operating temperature, and equipment operating status. By analyzing historical energy consumption data, energy consumption characteristic points during high energy consumption periods and under specific operating conditions are identified, and production data and equipment operating data corresponding to the energy consumption characteristic points are obtained, and the production data and equipment operating data corresponding to the energy consumption characteristic points are grouped.

[0097] In this embodiment, the energy consumption characteristic points correspond to key variables in the equipment operation process, such as changes in production speed, changes in equipment operating temperature, changes in equipment operating status, etc. Determining the energy consumption characteristic points can obtain the corresponding state changes of the equipment when the energy consumption changes at multiple time points, and then determine the detailed change data through the energy consumption changes at multiple time points, thereby establishing an energy consumption model.

[0098] In one embodiment, the energy consumption model is established as follows:

[0099] Obtain each set of data corresponding to the energy consumption characteristic point, build a random forest model for each set of data, and predict the energy consumption under the production conditions or equipment status;

[0100] Use the data set to optimize each model and optimize model parameters to improve prediction accuracy;

[0101] Cross-validation evaluates the performance of each model to avoid overfitting;

[0102] Combine multiple random forest models into an energy consumption model, and combine the prediction results of each model by weighted average or stacking;

[0103] According to the performance of each random forest model validation set, the weight of each random forest model is assigned to ensure that the prediction accuracy of the energy consumption model is maximized.

[0104] In one feasible method, multiple random forest models are combined to form an energy consumption model, which can accurately predict the energy consumption data based on the equipment status data and production data in different states, so as to determine the energy consumption that the equipment will generate. Based on the change in energy consumption, the energy consumption point that causes the energy consumption change can be determined, and the process and equipment operation can be optimized based on the energy consumption point data.

[0105] Among them, by combining multiple random forest models to assign different weights, the assigned weights of the random forest model can be adjusted in different environments, and the energy consumption prediction can be more targeted. For example, when the energy consumption point corresponds to production data, the weight of the random forest model A is 0.23, and when the energy consumption point corresponds to equipment status data, the weight of the random forest model A is 0.02.

[0106] In one embodiment, a validation set is used to evaluate the performance of each random forest model, and a weight is calculated based on the performance indicator of the random forest model:

[0107] Calculate the error of each model relative to the validation machine;

[0108] Calculate the inverse of the error as the preliminary weight, and normalize the preliminary weight so that its sum is 1;

[0109] The random forest model weights are calculated as follows:

[0110]

[0111] where w i is the weight of the i-th random forest model, MSE i is the mean square error of the ith random forest model, is the sum of the inverse errors of all random forest models;

[0112] Combine the predictions of each model using all random forest model weights;

[0113] Determine the final predicted value I of the energy consumption model;

[0114] I=w a I a +w b I b +…+w n I n ;

[0115] Among them, I n is the prediction result of the nth random forest model.

[0116] n represents the total number of models involved in the weighting. The prediction results of the random forest models will be combined together.

[0117] j is the index used to traverse all models. j takes values ​​from 1 to n, which is used to represent each model.

[0118] In one feasible method, since multiple groups of related production data and equipment operation data are determined based on energy consumption feature points, multiple random forest energy consumption models based on energy consumption feature points are generated, and the weights of multiple random forest energy consumption models are determined respectively. According to the prediction results of different weights, a final prediction value is generated, and the final prediction value corresponds to the prediction of the energy consumption model.

[0119] Exemplarily, three random forest models, random forest models A, B and C, are determined, and MSE is used as the performance indicator, as follows:

[0120] Model A: MSE A =0.02

[0121] Model B: MSE B =0.03

[0122] Model C: MSE C =0.01

[0123] Inverse error weighting method to calculate weights

[0124] Compute the inverse of the error:

[0125] Model A: 1 / MSE A =1 / 0.02=50

[0126] Model B: 1 / MSE B =1 / 0.03=33.33

[0127] Model C: 1 / MSE C =1 / 0.01=100

[0128] Normalize the initial weights:

[0129] Total: 50+33.33+100=183.33

[0130] Model A weight: 50 / 183.33≈0.273

[0131] Model B weight: 33.33 / 183.33≈0.182

[0132] Model C weight: 100 / 183.33≈0.545.

[0133] In one embodiment, the steps of obtaining device operating environment data and real-time device operating data and predicting changes in device operating data using the device operating environment data are as follows:

[0134] Determine the factors that affect the real-time equipment operation data and equipment operation environment data on the equipment operation data;

[0135] Calculate the impact of equipment operating environment data on equipment operating data;

[0136] Calculate the change range and change speed of equipment operation data based on the impact value;

[0137] Obtain the difference between the change of the equipment operation data in the subsequent time and the change amplitude and change speed of the calculated equipment operation data, and correct the equipment operation data of the predicted time according to the difference.

[0138] In this embodiment, by predicting the equipment operation data after a period of time, when the production data is the same, the change in the equipment operation data can be determined. According to the change in the equipment operation data and the fixed production data, the change in energy consumption caused by the equipment running to the predicted time can be calculated. According to the change in energy efficiency, the equipment operation state or equipment operation environment can be adjusted. Lower energy consumption can be achieved while ensuring stable production data of the equipment, thereby controlling energy consumption.

[0139] For example, the heat dissipation state of the device during operation is affected by the device environmental data. When the ambient temperature of the device is low, the device has a better heat dissipation state. When the ambient temperature of the device is high, the heating speed of the device will be accelerated, and it will reach a certain temperature earlier than at a low temperature.

[0140] In one embodiment, production data and equipment operation data of each process are obtained, and the total energy consumption of each process is obtained, and the energy consumption per unit output is calculated according to the production speed and output, and an energy consumption baseline of each process under normal conditions is established;

[0141] Determine the energy consumption overlap area of ​​the process, and use the energy consumption model to analyze the energy consumption overlap area of ​​the process, split the process steps, combine the process steps with overlapping areas and perform energy consumption simulation;

[0142] Use statistical methods and machine learning algorithms to identify energy consumption anomalies in energy consumption simulations;

[0143] Propose production optimization suggestions based on abnormal energy consumption analysis results.

[0144] In this embodiment, the process steps are divided into process step energy consumption and connection energy consumption. The process step energy consumption is the energy consumption necessary for the operation of the process step, and the connection energy consumption is the energy consumption between multiple steps of the process. The difference in connection energy consumption generated between multiple production lines is combined to determine whether the connection energy consumption between steps can be optimized.

[0145] After the process that can be optimized is optimized, energy consumption simulation is first performed to determine the energy consumption changes of the optimized process.

[0146] In one embodiment, the decision model performs simulations based on production optimization recommendations;

[0147] Design multiple optimization plans based on process flow and equipment operating status, including equipment maintenance, production scheduling, and process parameter adjustment;

[0148] Use the decision-making model to simulate each optimization plan and predict the changes in energy consumption, production efficiency, and cost indicators corresponding to each optimization plan;

[0149] Calculate the cost of equipment maintenance and production downtime based on the implementation details of each optimization plan;

[0150] Evaluate the expected benefits of each optimization solution: reduced energy consumption, increased production efficiency, and improved product quality;

[0151] Calculate the consumption and benefits of each optimization plan and determine the plan with the best comprehensive benefits.

[0152] In one feasible method, by determining the consumption and benefits of each optimization party, the cost and benefit relationship of adjusting the process or performing equipment maintenance can be calculated, so that the optimal plan for comprehensive benefits can be determined and decisions can be made based on the optimal plan for comprehensive benefits.

[0153] After determining the optimal solution, the external device is adjusted and controlled through the execution unit, and the execution result is fed back according to the change of the execution unit to the adjusted data. When performing subsequent optimization, the execution result is used as a reference.

[0154] In summary, the technical solutions disclosed in the above embodiments of the present invention have at least the following advantages:

[0155] 1. This application can make optimization suggestions for the production line based on the energy consumption of the production line by predicting the energy consumption of the production line. According to the energy consumption data corresponding to the operation data and production data of different equipment, the energy consumption hotspots can be determined, and the process steps can be optimized according to the energy consumption.

[0156] 2. By integrating data from different sources into a unified format, data collaboration across departments or systems can be carried out, and historical data can be stored and managed to facilitate traceability and long-term analysis. In combination with different levels of user access rights, data security can be protected.

[0157] 3. Through energy consumption models and decision-making models, different production optimization suggestions are simulated, and the advantages and disadvantages of each plan in terms of cost, energy consumption and efficiency are compared, so as to provide intelligent suggestions for production managers, such as adjusting the production line and changing the process, so as to achieve the goal of reducing costs and increasing efficiency.

[0158] 4. Collect feedback data after implementing the recommendations to adjust and improve the analysis model. Continuously optimize the model based on the feedback execution results to adapt to changes in the production environment and new data patterns. Monitor the operating data in real time. When anomalies or deviations from energy-saving targets are found, adjustments can be made through the execution unit.

[0159] Although the present invention is disclosed in conjunction with the above embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the attached claims.

Claims

1. A modeling and analysis system based on digital energy configuration, characterized in that: include: Data acquisition module, which obtains ERP data and MES data, and determines production data, energy consumption data, and equipment operation status data; The data cleaning module cleans the production data, energy consumption data and equipment operation status data collected by the data acquisition module, removes redundant data and determines data integrity, repairs missing values ​​according to the changes in the data before and after the missing data, and adjusts the data format standardization; The data integration module integrates different data, establishes a data source association network, and associates production data, energy consumption data, and equipment operation data; The data management module stores and backs up related data according to the data source association network and sets data access permissions to protect different types of data; The data preprocessing module determines the energy consumption characteristic points and groups the production data and equipment operation status data according to the energy consumption characteristic points; Energy consumption model: Establish energy consumption model based on production data, equipment operation data and equipment operation environment data grouped by energy consumption feature points. The energy consumption model reflects changes in production data and equipment operation data, and determines the impact of the equipment operation environment on the energy consumption model. Combined with changes in production data and equipment operation data, the equipment status is predicted and production optimization suggestions are made. Decision-making model, which makes optimization decisions based on the analysis of energy consumption points by the energy consumption model, determines the decision cost based on the decision, and calculates the energy consumption and efficiency of the optimization process; The execution unit controls external devices according to the optimization decision and obtains the data changes after the adjustment decision to feedback the execution result.

2. A modeling and analysis system based on digital energy configuration according to claim 1, characterized in that: ERP data and MES data include set data and real-time data. The set data is the production plan, and the real-time data is the changes in equipment data obtained through sensors installed on the equipment, as well as the energy consumption data of the equipment at different times.

3. A modeling and analysis system based on digital energy configuration according to claim 2, characterized in that: The data cleaning module performs data cleaning steps as follows: Compare key fields in production data, energy consumption data, and equipment operation data, identify duplicate data, and delete identical or redundant data records based on the identification results; Detect missing values, scan the data set, determine the location and number of missing values, and use interpolation algorithms to infer missing values ​​based on previous and subsequent data; Detect outliers in the data set through statistical analysis and correct them according to reasonable ranges or trends; Adjust the data format of the dataset so that the naming and format of the data fields are consistent.

4. A modeling and analysis system based on digital energy configuration according to claim 3, characterized in that: The steps to establish a data source association network are as follows: Determine the data source and clarify all data sources that need to be associated, including: production data, equipment operation data and energy consumption data, and determine the abnormal energy consumption points in the energy consumption data; Select appropriate associated fields, match energy consumption data with production data and equipment operation data through associated fields, and establish a data source association network; The data source association network is marked with a timestamp and stored according to the order of the timestamps.

5. A modeling and analysis system based on digital energy configuration according to claim 4, characterized in that: Identify and define key variables that can significantly affect energy consumption: production speed, equipment operating temperature, equipment operating status, and analyze historical energy consumption data to identify energy consumption characteristic points during high energy consumption periods and under specific working conditions, obtain the production data and equipment operation data corresponding to the energy consumption characteristic points, and group the production data and equipment operation data corresponding to the energy consumption characteristic points.

6. A modeling and analysis system based on digital energy configuration according to claim 5, characterized in that: The energy consumption model is established as follows: Obtain each set of data corresponding to the energy consumption characteristic point, build a random forest model for each set of data, and predict the energy consumption under the production conditions or equipment status; Use the data set to optimize each model and optimize model parameters to improve prediction accuracy; Cross-validation evaluates the performance of each model to avoid overfitting; Combine multiple random forest models into an energy consumption model, and combine the prediction results of each model by weighted average or stacking; According to the performance of each random forest model validation set, the weight of each random forest model is assigned to ensure that the prediction accuracy of the energy consumption model is maximized.

7. A modeling and analysis system based on digital energy configuration according to claim 6, characterized in that: Use the validation set to evaluate the performance of each random forest model and calculate the weights based on the performance metrics of the random forest model: Calculate the error of each model relative to the validation machine; Calculate the inverse of the error as the preliminary weight, and normalize the preliminary weight so that its sum is 1; The random forest model weights are calculated as follows: where w i is the weight of the i-th random forest model, MSE i is the mean square error of the ith random forest model, is the sum of the inverse errors of all random forest models; Combine the predictions of each model using all random forest model weights; Determine the final predicted value I of the energy consumption model; And=in a ·AND a +in b ·AND b +…+in n ·AND n ; Among them, I n is the prediction result of the nth random forest model.

8. The modeling and analysis system based on digital energy configuration according to claim 7 is characterized in that: The steps to obtain device operating environment data and real-time device operating data and predict changes in device operating data based on the device operating environment data are as follows: Determine the factors that affect the real-time equipment operation data and equipment operation environment data on the equipment operation data; Calculate the impact of equipment operating environment data on equipment operating data; Calculate the change range and speed of equipment operation data based on the impact value; Obtain the difference between the change of the equipment operation data in the subsequent time and the change amplitude and change speed of the calculated equipment operation data, and correct the equipment operation data of the predicted time according to the difference.

9. A modeling and analysis system based on digital energy configuration according to claim 8, characterized in that: Obtain the production data and equipment operation data of each process, and obtain the total energy consumption of each process, calculate the energy consumption per unit output based on the production speed and output, and establish the energy consumption baseline of each process under normal conditions; Determine the energy consumption overlap area of ​​the process, and use the energy consumption model to analyze the energy consumption overlap area of ​​the process, split the process steps, combine the process steps with overlapping areas and perform energy consumption simulation; Use statistical methods and machine learning algorithms to identify energy consumption anomalies in energy consumption simulations; Propose production optimization suggestions based on abnormal energy consumption analysis results.

10. A modeling and analysis system based on digital energy configuration according to claim 9, characterized in that: The decision model is simulated based on production optimization recommendations: Design multiple optimization plans based on process flow and equipment operating status, including equipment maintenance, production scheduling, and process parameter adjustment; Use the decision-making model to simulate each optimization plan and predict the changes in energy consumption, production efficiency, and cost indicators corresponding to each optimization plan; Calculate the cost of equipment maintenance and production downtime based on the implementation details of each optimization plan; Evaluate the expected benefits of each optimization solution: reduced energy consumption, increased production efficiency, and improved product quality; Calculate the consumption and benefits of each optimization plan and determine the plan with the best comprehensive benefits.

Images