A monitoring system for equipment operation process
By integrating high-precision sensors and machine learning algorithms into the equipment operation monitoring system, the problem of insufficient energy consumption optimization in existing technologies has been solved. This enables real-time monitoring of equipment energy consumption and customized energy-saving adjustments, thereby improving the energy management capabilities of enterprises.
Patent Information
- Application Number
- CN202411147215.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-21
AI Technical Summary
Existing equipment operation monitoring systems are inadequate in terms of energy consumption optimization, failing to achieve real-time energy consumption monitoring, in-depth analysis, and intelligent optimization, thus failing to meet the refined energy management needs of enterprises.
The equipment operation monitoring system, which integrates high-precision sensors and machine learning algorithms, includes an energy consumption data acquisition module, an energy consumption optimization model module, an energy-saving adjustment suggestion generation module, and a predictive maintenance module. It provides customized energy-saving adjustment suggestions by monitoring, analyzing, and predicting energy consumption trends in real time, and integrates with the enterprise's energy management system.
It enables real-time monitoring and in-depth analysis of equipment energy consumption, provides customized energy-saving adjustment suggestions, improves energy utilization efficiency and economic benefits, and ensures the accuracy and adaptability of the system.
Smart Images

Figure CN118656272B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of equipment monitoring, in particular to an equipment operation process monitoring system. Background Art
[0002] With the rapid development of global industrialization, energy consumption issues have increasingly become the focus of social attention. As an important means to improve production efficiency and reduce labor costs, industrial automation has been continuously applied in various fields. However, the energy consumption problem of automation equipment has gradually become prominent and has become a key factor restricting the sustainable development of enterprises. Energy costs account for an increasingly important proportion of enterprise operating costs. Therefore, effective energy consumption management is of great significance to reducing production costs and enhancing corporate competitiveness.
[0003] However, the current equipment operation process monitoring system has made certain progress in data collection and basic statistics, but it still has shortcomings in energy consumption optimization. Most of the existing systems can only provide a simple display of energy consumption data and lack in-depth energy consumption analysis and real-time optimization capabilities. This means that enterprises cannot dynamically adjust operating parameters according to the real-time operation of the equipment to achieve energy-saving purposes. In addition, the existing technology has limited capabilities in predicting equipment energy consumption patterns and intelligently recommending energy-saving measures, which makes it difficult to meet the needs of enterprises for refined energy management. Therefore, there is an urgent need for an equipment operation monitoring system that can achieve real-time energy consumption monitoring, in-depth analysis and intelligent optimization.
[0004] To sum up, the existing equipment operation process monitoring system has obvious defects in energy consumption optimization and cannot meet the growing energy management needs of enterprises. Therefore, it is of great practical significance and urgency to develop an equipment operation process monitoring system that can realize real-time energy consumption optimization and analysis. Summary of the invention
[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide an equipment operation process monitoring system, which can monitor the energy consumption of the equipment in real time, instantly analyze and predict energy consumption trends, identify energy consumption anomalies, and thus respond quickly and take measures. This real-time performance allows enterprises to respond immediately to high-energy consumption events, avoid unnecessary energy waste, and help with cost control and environmental responsibility fulfillment.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a device operation process monitoring system, the components of which include: an energy consumption data acquisition module, an energy consumption optimization model module, an energy-saving adjustment suggestion generation module, a predictive maintenance module, and an energy management integration module;
[0007] The energy consumption data acquisition module integrates high-precision current sensors, voltage sensors, power sensors, and temperature sensors to comprehensively capture the energy consumption and operating status information of the equipment;
[0008] The energy consumption optimization model module combines machine learning algorithms to build a multi-model fusion energy consumption optimization model, and provides customized solutions for energy consumption optimization under different working conditions by identifying different working conditions of the equipment, namely full load, light load, and no load;
[0009] The energy-saving adjustment suggestion generation module integrates an intelligent decision-making engine, and automatically generates energy-saving adjustment suggestions based on the prediction results of the energy consumption optimization model and the actual operation of the equipment. When generating energy-saving adjustment suggestions, multiple goals such as energy consumption reduction, production efficiency and equipment life are comprehensively considered;
[0010] The predictive maintenance module combines equipment operation data and historical fault records to predict equipment failures and issue early warnings. Based on the fault warning results, it automatically generates a maintenance plan for the equipment, including maintenance time, maintenance content and required spare parts, performs performance evaluation and energy consumption analysis on the equipment after maintenance, verifies the maintenance effect, and continuously optimizes the maintenance plan.
[0011] The energy management integration module is integrated with other enterprise energy management systems, and realizes data interconnection and interoperability through standard interfaces, so that energy consumption data can flow freely between various systems within the enterprise. Based on the integrated data, the module can analyze the overall energy usage of the enterprise, and provide the enterprise with comprehensive energy strategy optimization suggestions in combination with production plans and business needs.
[0012] Furthermore, the energy consumption data collection module also includes:
[0013] Data collection frequency setting submodule: Determine the data collection frequency based on the operating characteristics of the equipment and the speed of energy consumption changes. Use adaptive sampling technology to automatically increase the collection frequency when the equipment is in operation and when energy consumption fluctuates abnormally, so as to capture more detailed energy consumption change information.
[0014] Data synchronization and calibration submodule: synchronize the data collected by different types of sensors in time to accurately reflect the energy consumption status of the equipment at the same time, and calibrate the collected data to eliminate the measurement error of the sensor. A linear calibration algorithm is used, i.e. ,in is the original measurement value, and is the coefficient obtained by calibration with a standard source, is the accurate value after calibration.
[0015] Furthermore, the data acquisition frequency setting submodule analyzes the statistical characteristics of the equipment energy consumption data, identifies the key time points and conditions of energy consumption changes, namely, equipment startup and load changes, and dynamically sets the adaptive sampling frequency adjustment according to the changing characteristics of the energy consumption data and the key acquisition points, that is, the calculation process is: , is the sampling frequency, is the energy consumption value of the device collected at the current moment, is the average energy consumption, i.e. , is the number of samples for calculating the average energy consumption, k and It is the adjustment coefficient that continuously optimizes the acquisition frequency setting based on historical data and real-time feedback, so that the system can automatically adjust the acquisition frequency to adapt to different operating conditions.
[0016] Furthermore, the energy consumption optimization model module is combined with a machine learning algorithm to construct a hierarchical model. The first layer identifies the operating conditions, i.e., full load, light load, and no-load. The second layer selects an energy consumption prediction model according to the operating conditions, and classifies the operating conditions of the equipment according to the equipment's operating speed, load, current parameters, temperature, and vibration state information. Features related to operating condition identification are extracted from the original data, and the training data set is labeled to clarify the operating condition category corresponding to each sample, i.e., full load, light load, and no-load. Multiple energy consumption prediction models are pre-trained for each operating condition and stored in the model library. According to the operating conditions identified by the first layer, the corresponding energy consumption prediction model is selected from the model library, and the selected model is used to predict energy consumption. The accuracy of the prediction results is evaluated according to actual needs. The accuracy of the model is evaluated by a cross-validation method, and the trained model is deployed to the monitoring system to receive new data in real time and identify the operating conditions.
[0017] Furthermore, the energy consumption optimization model module establishes an energy consumption prediction model for the equipment based on the extracted energy consumption characteristics under different operating states and loads, and the model building process is as follows:
[0018] Data preparation: Collect a large amount of energy consumption characteristic data under different operating states and loads of the equipment to form a training data set. ,in , including operating status ,load , Operation parameters , represents the corresponding actual energy consumption value, i=1,2,···,n;
[0019] Select the decision tree algorithm: The decision tree algorithm selects the optimal features for node splitting based on information gain, that is, , is the current dataset, It is a feature, It is a feature All values of It is a feature The value is The corresponding subset is is the entropy of the dataset, Yes Category In the dataset The proportion of
[0020] Construct a decision tree: Starting from the root node, select the feature with the largest information gain as the split feature, divide the data set into multiple subsets according to the value of the feature, repeat the above steps for each subset, and recursively construct a subtree until the stopping condition is met, that is, the number of samples in the node is less than the threshold;
[0021] Model evaluation and optimization: The constructed decision tree model is evaluated using the test data set using the mean absolute error method, that is, , It is the predicted value of the model. According to the evaluation results, the model is adjusted and optimized, that is, the depth of the tree is adjusted and the branches are pruned;
[0022] Final model: The final decision tree energy consumption prediction model is obtained after optimization.
[0023] Furthermore, the energy consumption optimization model uses sensors installed on the equipment to collect energy consumption data of current, voltage and power of the equipment at a predetermined frequency, divides the data into a training set, a validation set and a test set, and establishes an energy consumption prediction model of the equipment based on the extracted energy consumption characteristics under different operating states and loads. The model is trained using a decision tree algorithm, and the final decision tree model formula is: , represents the predicted energy consumption, Indicates the operating status of the device. Indicates the equipment load, Indicates the equipment operating parameters. is the error term, which is the part that cannot be explained by the model, is the intercept term, , , , … are the weight coefficients of each feature. A large amount of historical energy consumption data is used for iterative optimization. The prediction error of the model is minimized by adjusting the weight coefficients. When the performance on the validation set meets the accuracy requirements and convergence conditions, the training is completed.
[0024] Furthermore, the energy-saving adjustment suggestion generation module calculates the difference between the current energy consumption and the predicted optimal energy consumption by comparing the real-time energy consumption data with the optimal energy consumption value output by the energy consumption optimization model, and analyzes the causes of the difference, namely, the equipment operation status, environmental parameters, and load changes. The analysis process is as follows:
[0025] Data integration and model prediction: Integrate the prediction results from the energy consumption optimization model module, integrate the real-time equipment operation status, operating parameters, energy consumption data and predicted energy consumption values;
[0026] Multi-objective optimization problem: define the optimization objectives, i.e., minimize energy consumption, maximize production efficiency, and maintain equipment life, convert each objective into a quantifiable indicator, and set corresponding weights to reflect the importance the company attaches to these objectives;
[0027] Energy-saving strategy generation: Based on the multi-objective optimization problem, generate preliminary energy-saving adjustment strategies, including adjusting equipment operating parameters, i.e., speed, temperature set point, load distribution, changing the operating mode, i.e., enabling energy-saving mode, adjusting working hours, and evaluating the impact of preliminary strategies on energy consumption, production efficiency, and equipment life;
[0028] Strategy optimization and selection: Based on the evaluation results, the preliminary strategy is optimized, that is, parameter values are adjusted, different strategies are combined, and the best energy-saving adjustment suggestion is selected, considering its feasibility, economic benefits and implementation difficulty.
[0029] Furthermore, the predictive maintenance module uses historical fault records to classify and mark faults to form a fault mode library. Based on the extracted features and marked fault modes, a fault prediction model is constructed using time series analysis, i.e. , Indicates at time The observed value of is a constant term, , ,……, are model parameters, is the order of the autoregressive model, It is white noise. Use historical data to train the model, estimate model parameters, check the properties of model residuals to ensure that they are white noise, use a validation set outside the training data to test the predictive ability of the model, use the model to predict the operating parameters of the future equipment, and trigger a fault warning when the predicted value exceeds the preset threshold.
[0030] Furthermore, when multiple related devices work together, the energy management module collects the operating parameters and energy consumption data of each device in real time. The operating parameters include but are not limited to the operating frequency, output power, and operating status of the device. The energy consumption data includes real-time power consumption and cumulative energy consumption. The operation matching degree between these devices is analyzed. For example, for the front and rear process equipment on a production line, their production speed and capacity are compared, and the matching index is calculated. ,Right now , Represents the output of the preceding device. Represents the input amount of subsequent equipment. The closer the ratio is to 1, the higher the matching degree. At the same time, it evaluates the collaborative efficiency between equipment and measures the collaborative efficiency by calculating the ratio of the total energy consumption of the entire collaborative work system to the total output. ,Right now , represents the total energy consumption, Represents total output. The lower the ratio, the higher the synergy efficiency.
[0031] Furthermore, the energy management module establishes an energy consumption optimization model based on multi-device collaboration according to the analysis of multi-device matching degree and collaboration rate, that is, the operating parameters of the devices are , the collaborative relationship between devices is represented by a matrix The model formula is: , Represents the impact function of a single device operating parameter on energy consumption, It represents the influence function of the collaborative relationship between devices on energy consumption. By finding the combination of equipment operating parameters and collaborative relationship configuration that minimizes energy consumption, and using the optimization results of the model, an energy-saving optimization plan for the collaborative operation of multiple devices is given. The plan includes adjusting the working frequency, startup sequence, and operating time allocation of the equipment to achieve overall energy consumption reduction and improved collaborative efficiency. It continuously monitors the actual operation of the equipment, compares the actual data with the model prediction results, and constantly corrects and improves the energy consumption optimization model to adapt to changes in equipment operating conditions and work tasks.
[0032] Compared with the prior art, this equipment operation process monitoring system has the following beneficial effects:
[0033] 1. The present invention realizes real-time monitoring and in-depth analysis of energy consumption during equipment operation. By integrating high-precision sensors, the system can comprehensively capture the energy consumption and operating status information of the equipment, including current, voltage, power, temperature and vibration parameters. This real-time monitoring can not only detect abnormal energy consumption in a timely manner, but also monitor the operating status of the equipment in detail, thereby providing accurate data support for energy consumption optimization.
[0034] 2. The present invention combines machine learning algorithms to construct an energy consumption optimization model that integrates multiple models, which can provide customized energy consumption optimization solutions according to different working conditions. At the same time, the energy-saving adjustment suggestion generation module can automatically generate targeted and operational energy-saving adjustment suggestions based on the prediction results of the energy consumption optimization model and the actual operation of the equipment, thereby effectively reducing energy consumption and improving the energy utilization efficiency and economic benefits of the enterprise. In addition, the real-time feedback and optimization mechanism can continuously improve the energy consumption optimization model and the energy-saving adjustment suggestion generation algorithm, and improve the accuracy and adaptability of the system.
[0035] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0037] Figure 1 The operational flow chart of a device operation process monitoring system.
[0038] Figure 2 Flowchart of a process monitoring system for equipment operation. DETAILED DESCRIPTION
[0039] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0040] Embodiment 1
[0041] like Figure 1 , Figure 2 As shown, this embodiment provides an equipment operation process monitoring system, which can capture the energy consumption and operation status information of the equipment in real time by integrating high-precision sensors and advanced data processing and analysis technologies, conduct in-depth analysis, and promptly discover abnormal energy consumption. This real-time performance enables enterprises to respond quickly and take measures to avoid energy waste and improve energy utilization efficiency.
[0042] In the specific implementation, the system first installs high-precision current sensors, voltage sensors, power sensors, temperature sensors and vibration sensors on the key equipment of each production line to ensure that all sensors can work stably and connect with the data acquisition module of the monitoring system, including energy consumption data acquisition module, energy consumption optimization model module, energy-saving adjustment suggestion generation module, predictive maintenance module and energy management integration module, configure system parameters to ensure that each module can interact and share data normally, and collect energy consumption data, operating status and operating parameters of the equipment at a preset frequency through the sensors installed on the equipment. The data includes the equipment's power-on, standby, running, and shutdown status, speed, load and other operating parameters, as well as the corresponding current, voltage, power and other energy consumption values. The collected raw data is cleaned, and the data synchronization and calibration submodule ensures that the data collected by different sensors are synchronized in time, and uses a linear calibration algorithm to calibrate the data, that is, ,The data acquisition frequency setting submodule analyzes the statistical characteristics of the ,equipment energy consumption data, identifies the key time points and conditions of ,energy consumption changes, i.e., equipment startup and load changes, and ,dynamically sets the adaptive sampling frequency adjustment according to the ,changing characteristics of the energy consumption data and the key ,acquisition points, i.e. , is the sampling frequency, E dqnh is the energy consumption value of the device collected at the current moment, is the average energy consumption, i.e. , extract features from it, such as energy consumption characteristics under different operating states, energy consumption change trends under different loads, etc., and store the processed data in the system database for subsequent analysis and modeling.
[0043] Then, the energy consumption optimization model module combines the machine learning algorithm to build a hierarchical model, and classifies the equipment's operating conditions into full load, light load, and no-load according to the equipment's operating speed, load, current parameters, temperature, and vibration status information. Features related to operating condition identification are extracted from the original data, and the training data set is labeled to clarify the operating condition category corresponding to each sample. Multiple energy consumption prediction models are pre-trained for each operating condition and stored in the model library. According to the operating condition identified by the first layer, the corresponding energy consumption prediction model is selected from the model library for energy consumption prediction. When it is judged that it is in a full-load condition, the corresponding full-load condition energy consumption prediction model is selected. The model is based on the decision tree algorithm, and the input parameters include the speed, load, temperature, etc. of the injection molding machine, and the predicted energy consumption value is output.
[0044] Next, the energy-saving adjustment suggestion generation module calculates the difference between the current energy consumption and the predicted optimal energy consumption by comparing the real-time energy consumption data with the optimal energy consumption value output by the energy consumption optimization model, and analyzes the reasons for the difference, such as equipment operating status, environmental parameters, load changes, etc. If it is found that the current energy consumption of a certain device is too high due to unreasonable operating parameters, the system will generate suggestions for adjusting the equipment speed and temperature set points, define the optimization goals as minimizing energy consumption, maximizing production efficiency, and maintaining equipment life, and set corresponding weights. Based on the multi-objective optimization problem, it generates preliminary energy-saving adjustment strategies, such as enabling energy-saving mode or adjusting working hours, and evaluates the impact on energy consumption, production efficiency and equipment life. According to the evaluation results, the preliminary strategy is optimized and the optimal energy-saving adjustment suggestion is selected. For multiple devices on a production line, the conditions of each device are comprehensively considered, and the operating frequency and load distribution of some devices are adjusted to achieve overall energy consumption reduction and improved production efficiency.
[0045] Secondly, the predictive maintenance module uses historical fault records to classify and mark faults to form a fault mode library. Based on the extracted features and marked fault modes, a fault prediction model is constructed using time series analysis, i.e. , the model parameters are obtained through training based on historical data. When the operating parameters monitored in real time are input into the model, if the predicted value exceeds the preset threshold, the system triggers a fault warning. According to the fault warning result, the maintenance plan of the equipment is automatically generated, including maintenance time, maintenance content and required spare parts. After maintenance, the equipment is evaluated for performance and analyzed for energy consumption to verify the maintenance effect, and the maintenance plan is continuously optimized.
[0046] Finally, the equipment operating parameters and energy consumption data are collected in real time, the operation matching and collaborative efficiency between the equipment are analyzed, the coordination status between the equipment is evaluated by calculating the matching index and the collaborative efficiency ratio, and an energy consumption optimization model based on the equipment operating parameters and collaborative relationship is established. The energy-saving plan is given through the model optimization results, including the adjustment of the equipment operating frequency, startup sequence and operating time allocation, and the model is continuously monitored and corrected.
[0047] In summary, this embodiment describes in detail the various modules of a device operation process monitoring system and its implementation process, demonstrating that the system's energy consumption optimization model based on machine learning can provide customized energy-saving adjustment suggestions according to different working conditions. In addition, the system continuously optimizes the model and algorithm through user feedback and continuous monitoring of actual energy consumption changes of the equipment, thereby ensuring the accuracy and applicability of the energy-saving suggestions, thereby improving the performance of the entire system and the energy utilization efficiency of the equipment.
[0048] Embodiment 2
[0049] A device operation process monitoring system in this embodiment particularly emphasizes providing customized energy consumption optimization solutions for different operating conditions (full load, light load, and no-load) of equipment through machine learning algorithms, and integrating with other enterprise energy management systems to achieve data interconnection and interoperability, thereby being able to analyze the overall energy usage of the enterprise, and combining production plans and business needs to provide the enterprise with comprehensive energy strategy optimization suggestions.
[0050] First, according to the equipment's operating speed, load, current parameters, temperature, and vibration status information, ,in , including operating status ,load , Operation parameters , Represents the corresponding actual energy consumption value, i=1,2,···,n. The energy consumption optimization model module divides the working conditions of the equipment into full load, light load, and no load. Features related to working condition identification are extracted from the original data, such as the equipment's operating speed, load size, current fluctuation range, temperature mean, and vibration amplitude. The training data set is labeled to clarify the working condition category corresponding to each sample. Multiple energy consumption prediction models are pre-trained for each working condition and stored in the model library. These models use machine learning algorithms to select the optimal features for node splitting based on information gain, i.e. , is the entropy of the data set. In actual operation, according to the real-time collected equipment operation data, the current working condition is first identified, and then the corresponding energy consumption prediction model is selected from the model library. The mean absolute error method is used to evaluate the constructed decision tree model using the test data set, that is, ,The trained model is deployed to the monitoring system, enabling it to receive new data in real time and perform working condition identification and energy consumption prediction.
[0051] The energy-saving adjustment suggestion generation module integrates the prediction results from the energy consumption optimization model module, and integrates the real-time equipment operation status, operating parameters, energy consumption data and predicted energy consumption values. It defines the optimization goals as minimizing energy consumption, maximizing production efficiency and maintaining equipment life, and converts these goals into quantifiable indicators, such as the percentage of energy consumption reduction, the improvement in production efficiency and the increase in the remaining life of the equipment.
[0052] For scenarios where multiple related devices work together, the energy management integration module collects the operating parameters and energy consumption data of each device in real time, including operating frequency, output power, operating status, real-time power consumption and cumulative energy consumption, and analyzes the operating matching indicators between these devices. ,Right now At the same time, the collaborative efficiency between devices is evaluated by calculating the ratio of the total energy consumption of the entire collaborative work system to the total output. ,Right now According to the analysis of multi-device matching degree and coordination rate, an energy consumption optimization model based on multi-device coordination is established, that is, the operating parameters of the equipment are , the collaborative relationship between devices is represented by a matrix The model formula is: ,By finding the equipment operating parameter combination and synergistic relationship configuration that minimizes energy consumption, and using the optimization results of the model, an energy-saving optimization plan for the coordinated operation of multiple devices is given, that is, adjusting the operating frequency, startup sequence, and operating time allocation of the equipment.
[0053] Finally, continuously monitor the actual operation of the equipment, compare the actual data with the model prediction results, and continuously revise and improve the energy consumption optimization model to adapt to changes in equipment operating conditions and work tasks.
[0054] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. A device operation process monitoring system, It is characterized in that The components of the monitoring system include: energy consumption data acquisition module, energy consumption optimization model module, energy-saving adjustment suggestion generation module, predictive maintenance module, and energy management integration module; The energy consumption data acquisition module integrates high-precision current sensors, voltage sensors, power sensors, and temperature sensors to comprehensively capture the energy consumption and operating status information of the equipment; The energy consumption optimization model module combines machine learning algorithms to build a multi-model fusion energy consumption optimization model, and provides customized solutions for energy consumption optimization under different working conditions by identifying different working conditions of the equipment, including full load, light load, and no load; The energy-saving adjustment suggestion generation module integrates an intelligent decision-making engine, and automatically generates energy-saving adjustment suggestions based on the prediction results of the energy consumption optimization model and the actual operation of the equipment. When generating energy-saving adjustment suggestions, multiple goals such as energy consumption reduction, production efficiency and equipment life are comprehensively considered; The predictive maintenance module combines equipment operation data and historical fault records to predict equipment failures and issue early warnings. Based on the fault warning results, it automatically generates a maintenance plan for the equipment, including maintenance time, maintenance content and required spare parts, performs performance evaluation and energy consumption analysis on the equipment after maintenance, verifies the maintenance effect, and continuously optimizes the maintenance plan. The energy management integration module is integrated with other enterprise energy management systems to achieve data interconnection and intercommunication through standard interfaces, so that energy consumption data can flow freely between various systems within the enterprise. Based on the integrated data, the module can analyze the overall energy usage of the enterprise, and provide comprehensive energy strategy optimization suggestions for the enterprise in combination with production plans and business needs; The energy consumption data collection module also includes: Data collection frequency setting submodule: Determine the data collection frequency based on the operating characteristics of the equipment and the speed of energy consumption changes. Use adaptive sampling technology to automatically increase the collection frequency when the equipment is in operation and when energy consumption fluctuates abnormally, so as to capture more detailed energy consumption change information. Data synchronization and calibration submodule: synchronizes the data collected by different types of sensors in time to accurately reflect the energy consumption status of the equipment at the same time, and calibrates the collected data to eliminate the measurement error of the sensor. The linear calibration algorithm is used. ,in is the original measurement value, and is the coefficient obtained by calibration with a standard source, is the accurate value after calibration; The data acquisition frequency setting submodule analyzes the statistical characteristics of the equipment energy consumption data, identifies the key time points and conditions of energy consumption changes, and the key time points of energy consumption changes are the equipment startup point and the load change point. According to the change characteristics of the energy consumption data and the key acquisition points, the adaptive sampling frequency adjustment is dynamically set. The calculation process is: , is the sampling frequency, is the energy consumption value of the device collected at the current moment, is the average energy consumption, the average energy consumption , is the number of samples for calculating the average energy consumption, and It is the adjustment coefficient that continuously optimizes the acquisition frequency setting based on historical data and real-time feedback, so that the system can automatically adjust the acquisition frequency to adapt to different operating conditions.
2. A device operation process monitoring system according to claim 1, It is characterized in that The energy consumption optimization model module is combined with a machine learning algorithm to construct a hierarchical model. The first layer identifies the working condition, and the second layer selects an energy consumption prediction model according to the working condition. The working condition of the equipment is classified according to the equipment's operating speed, load, current parameters, temperature, and vibration state information. Features related to working condition identification are extracted from the original data, and the training data set is labeled to clarify the working condition category corresponding to each sample. Multiple energy consumption prediction models are pre-trained for each working condition and stored in a model library. According to the working condition identified by the first layer, the corresponding energy consumption prediction model is selected from the model library, and energy consumption is predicted using the selected model. The accuracy of the prediction result is evaluated according to actual needs. The accuracy of the model is evaluated through a cross-validation method, and the trained model is deployed to the monitoring system to receive new data in real time and identify the working condition.
3. A device operation process monitoring system according to claim 2, It is characterized in that The energy consumption optimization model module establishes an energy consumption prediction model for the equipment based on the extracted energy consumption characteristics under different operating states and loads. The model building process is as follows: Data preparation: Collect a large amount of energy consumption characteristic data under different operating states and loads of the equipment to form a training data set. ,in , including operating status ,load , Operation parameters , represents the corresponding actual energy consumption value, i=1,2,···,n; Select the decision tree algorithm: The decision tree algorithm selects the optimal features for node splitting based on information gain, specifically: , is the current dataset, It is a feature, It is a feature All values of It is a feature The value is The corresponding subset is is the entropy of the dataset, Yes Category In the dataset The proportion of Construct a decision tree: Starting from the root node, select the feature with the largest information gain as the split feature, divide the data set into multiple subsets according to the value of the feature, repeat the above steps for each subset, and recursively construct a subtree until the stopping condition is met. The stopping condition is: the number of samples in the node is less than the threshold; Model evaluation and optimization: Use the mean absolute error method to evaluate the constructed decision tree model using the test data set, specifically: , is the predicted value of the model. According to the evaluation result, the model is adjusted and optimized. The adjustment and optimization include: adjusting the depth of the tree and pruning; Final model: The final decision tree energy consumption prediction model is obtained after optimization.
4. A device operation process monitoring system according to claim 3, It is characterized in that The energy consumption optimization model uses sensors installed on the equipment to collect the energy consumption data of the equipment's current, voltage and power at a predetermined frequency, divides the data into a training set, a validation set and a test set, and establishes an energy consumption prediction model for the equipment based on the extracted energy consumption characteristics under different operating states and loads. The model is trained using a decision tree algorithm, and the final decision tree model formula is: , represents the predicted energy consumption, Indicates the operating status of the device. Indicates the equipment load, Indicates the equipment operating parameters. is the error term, which is the part that cannot be explained by the model, is the intercept term, , , , … are the weight coefficients of each feature. A large amount of historical energy consumption data is used for iterative optimization. The prediction error of the model is minimized by adjusting the weight coefficients. When the performance on the validation set meets the accuracy requirements and convergence conditions, the training is completed.
5. The equipment operation process monitoring system according to claim 1, It is characterized in that The energy-saving adjustment suggestion generation module calculates the difference between the current energy consumption and the predicted optimal energy consumption by comparing the real-time energy consumption data with the optimal energy consumption value output by the energy consumption optimization model, and analyzes the reasons for the difference, including analyzing the equipment operation status, environmental parameters, and load changes. The analysis process is as follows: Data integration and model prediction: Integrate the prediction results from the energy consumption optimization model module, integrate the real-time equipment operation status, operating parameters, energy consumption data and predicted energy consumption values; Multi-objective optimization problem: define optimization goals, including minimizing energy consumption, maximizing production efficiency, and maintaining equipment life, convert each goal into a quantifiable indicator, and set corresponding weights to reflect the importance the company attaches to these goals; Energy-saving strategy generation: Based on the multi-objective optimization problem, generate preliminary energy-saving adjustment strategies, including adjusting equipment operating parameters and changing operating modes. The equipment operating parameters include: speed, temperature set point, load distribution. The operating modes include: enabling energy-saving mode and adjusting working hours. Evaluate the impact of preliminary strategies on energy consumption, production efficiency and equipment life. Strategy optimization and selection: Based on the evaluation results, the preliminary strategy is optimized. The optimization of the preliminary strategy includes: adjusting parameter values, combining different strategies, selecting the best energy-saving adjustment suggestion, and considering its feasibility, economic benefits and implementation difficulty.
6. The equipment operation process monitoring system according to claim 1, It is characterized in that The predictive maintenance module uses historical fault records to classify and mark faults to form a fault mode library. Based on the extracted features and marked fault modes, a time series analysis algorithm is used to build a fault prediction model, specifically: , Indicates at time The observed value of is a constant term, , ,……, are model parameters, is the order of the autoregressive model, It is white noise. Use historical data to train the model, estimate model parameters, check the properties of model residuals to ensure that they are white noise, use a validation set outside the training data to test the predictive ability of the model, use the model to predict the operating parameters of the future equipment, and trigger a fault warning when the predicted value exceeds the preset threshold.
7. The equipment operation process monitoring system according to claim 1, It is characterized in that When multiple related devices work together, the energy management integrated module collects the operating parameters and energy consumption data of each device in real time. The operating parameters include but are not limited to the operating frequency, output power, and operating status of the device. The energy consumption data includes real-time power consumption and cumulative energy consumption. The operation matching degree between these devices is analyzed; for the front and rear process equipment on a production line, their production speed and capacity are compared, and the matching index is calculated. , matching index , Represents the output of the preceding device. Represents the input amount of subsequent equipment. The closer the ratio is to 1, the higher the matching degree. At the same time, it evaluates the collaborative efficiency between equipment and measures the collaborative efficiency by calculating the ratio of the total energy consumption of the entire collaborative work system to the total output. , collaborative efficiency , represents the total energy consumption, Represents total output. The lower the ratio, the higher the synergy efficiency.
8. A device operation process monitoring system according to claim 7, It is characterized in that The energy management integration module establishes an energy consumption optimization model based on multi-device collaboration according to the analysis of multi-device matching degree and collaboration rate. If the operating parameters of the devices are , the collaborative relationship between devices is represented by a matrix The model formula is: , Represents the impact function of a single device operating parameter on energy consumption, It represents the influence function of the collaborative relationship between devices on energy consumption. By finding the combination of equipment operating parameters and collaborative relationship configuration that minimizes energy consumption, and using the optimization results of the model, an energy-saving optimization plan for the collaborative operation of multiple devices is given. The plan includes adjusting the working frequency, startup sequence, and operating time allocation of the equipment to achieve overall energy consumption reduction and improved collaborative efficiency. It continuously monitors the actual operation of the equipment, compares the actual data with the model prediction results, and constantly corrects and improves the energy consumption optimization model to adapt to changes in equipment operating conditions and work tasks.
Citation Information
Cited By
Continuous production intelligent control system based on precipitation enhancement and hail suppression rocket projectile delay powder
CN122569225A
