CN117874464A - Vertical mill energy consumption optimization method based on deep learning and data mining - Google Patents
Vertical mill energy consumption optimization method based on deep learning and data mining Download PDFInfo
- Publication number
- CN117874464A CN117874464A CN202410045317.0A CN202410045317A CN117874464A CN 117874464 A CN117874464 A CN 117874464A CN 202410045317 A CN202410045317 A CN 202410045317A CN 117874464 A CN117874464 A CN 117874464A
- Authority
- CN
- China
- Prior art keywords
- data
- energy consumption
- deep learning
- vertical mill
- data mining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 62
- 238000005457 optimization Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000007418 data mining Methods 0.000 title claims abstract description 42
- 238000013135 deep learning Methods 0.000 title claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000013136 deep learning model Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 238000005065 mining Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 2
- 125000004122 cyclic group Chemical group 0.000 claims description 2
- 238000000227 grinding Methods 0.000 abstract description 5
- 238000009412 basement excavation Methods 0.000 abstract description 4
- 238000007405 data analysis Methods 0.000 abstract description 2
- 239000000463 material Substances 0.000 description 7
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 6
- 239000004568 cement Substances 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000003245 coal Substances 0.000 description 2
- 239000000843 powder Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012824 chemical production Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000012407 engineering method Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a vertical mill energy consumption optimization method based on deep learning and data mining, which comprises the following steps: step 1, collecting operation data of vertical mill equipment, and preprocessing the data; training the deep learning model by using the data preprocessed in the step 1, and applying the trained model to energy consumption result prediction; step 3, performing data mining analysis by using the data preprocessed in the step 1 to acquire association rules; and 4, generating an energy consumption optimization strategy by combining the prediction result of the deep learning model in the step 2 and the association rule of the data mining analysis in the step 3, and operating the vertical mill equipment by using the energy consumption optimization strategy. The invention can perform data analysis and excavation on the grinding equipment more deeply and comprehensively, and realizes more accurate and flexible energy consumption optimization control by combining deep learning.
Description
Technical Field
The invention relates to the technical field of vertical mill control, in particular to a vertical mill energy consumption optimization method based on deep learning and data mining.
Background
The application of the vertical mill equipment is wide: vertical mill equipment has found wide application in a variety of industrial processes including, but not limited to, cement manufacture, ore processing, chemical production, and the like. These devices typically require long, high load operation, so energy consumption issues have been the focus of optimization. The vertical mill equipment occupies an important position in industrial production, but how to improve the operation efficiency and reduce the energy consumption thereof has long been the core subject of research and optimization. Traditional data acquisition and processing methods rely primarily on physical sensors to acquire device parameters, but these sensors may have difficulty measuring all critical parameters directly in some situations. In addition, existing methods often have difficulty meeting the stringent requirements for real-time, efficient and accurate data. Meanwhile, the existing optimization method is mostly dependent on experience and a simple mathematical model, so that accuracy and comprehensiveness are lacking.
Disclosure of Invention
The invention aims to provide a vertical mill energy consumption optimization method based on deep learning and data mining. The invention can perform data analysis and excavation on the grinding equipment more deeply and comprehensively, and realizes more accurate and flexible energy consumption optimization control by combining deep learning.
The technical scheme of the invention is as follows: the vertical mill energy consumption optimization method based on deep learning and data mining comprises the following steps:
step 1, collecting operation data of vertical mill equipment, and preprocessing the data;
training the deep learning model by using the data preprocessed in the step 1, and applying the trained model to energy consumption result prediction;
step 3, performing data mining analysis by using the data preprocessed in the step 1 to acquire association rules;
and 4, generating an energy consumption optimization strategy by combining the prediction result of the deep learning model in the step 2 and the association rule of the data mining analysis in the step 3, and operating the vertical mill equipment by using the energy consumption optimization strategy.
In the above vertical mill energy consumption optimization method based on deep learning and data mining, in step 1, the preprocessing of data includes removing abnormal values and noise by using a data cleaning method, then performing data standardization and normalization, and then performing discretization processing on the data.
The discrete processing adopts state cluster analysis; the state clustering analysis is based on a Kmeans algorithm, data to be clustered is selected, each column in the data represents an attribute, each row represents a record of a working condition state, then the number of clusters required by the clustering and the iteration number are set, when the clustering center is not changed or the maximum iteration number is reached, the clustering is stopped, and then the category labeling is completed on the original clustered data.
In the foregoing vertical mill energy consumption optimization method based on deep learning and data mining, in step 2, the deep learning model is a convolutional neural network, a cyclic neural network and/or a long-short-term memory network.
In the foregoing vertical mill energy consumption optimization method based on deep learning and data mining, in step 3, discretized data is used as input data, and association rule mining is performed based on Apriori algorithm by setting minimum support and minimum confidence parameters.
According to the vertical mill energy consumption optimization method based on deep learning and data mining, the support degree is calculated as follows:
the confidence is calculated as follows:
according to the vertical mill energy consumption optimization method based on deep learning and data mining, the output result of the evaluation function of the Apriori algorithm is a rule between an adjustable parameter and a stable index or between the adjustable parameter and an energy consumption index meeting the requirements of minimum support and minimum confidence, and the rule is calculated as follows:
wherein: f (x) is the original evaluation function,the representation rule includes interesting or valid rules.
The system of the vertical mill energy consumption optimization method based on deep learning and data mining comprises a data acquisition module, a data preprocessing module, a data mining module and an optimization strategy generation module;
the data acquisition module acquires data of the vertical face equipment by using various sensors;
the data preprocessing module applies an algorithm to perform data preprocessing;
the data mining module uses a data mining algorithm to analyze real-time data;
and the optimization strategy generation module generates energy consumption optimization strategies and rules according to the data mining result.
The system of the vertical mill energy consumption optimization method based on deep learning and data mining further comprises a virtual sensor module, wherein the virtual sensor module is used for simulating and predicting certain equipment parameters which cannot be directly measured through an algorithm.
Compared with the prior art, the method not only adopts advanced data acquisition and mining algorithm, but also integrates a deep learning model, ensures more efficient and accurate recognition of the energy consumption mode and automatically generates an optimization strategy. The method combining data mining and deep learning overcomes the limitation of the traditional optimization method, and can perform intelligent, accurate and flexible optimization on the energy consumption of the counter grinding equipment. The data acquisition of the present invention aims to provide real-time, efficient and accurate operational data for a vertical mill apparatus. The invention not only can collect the operation condition data of the vertical mill equipment, but also can ensure the accuracy and the real-time performance of the data through a data processing algorithm, and provides solid data support for the operation optimization, the energy consumption monitoring and the fault prediction of the vertical mill equipment. The present invention further comprises the following
The beneficial effects are that:
1. adaptivity: by combining deep learning, data mining and reinforcement learning, the invention can analyze and decide the running state of the equipment in real time, dynamically adapt to various working condition changes and ensure that the equipment always runs in an optimal or near-optimal state.
2. Real-time and intelligent: the method not only improves the instantaneity of the optimization strategy, but also greatly improves the intelligence of the strategy, so that the optimization is more timely and accurate.
3. Deep feature learning: the method can capture deep features in the original data, and is more efficient and accurate than the traditional feature engineering method.
4. Advanced data processing and mining techniques: the invention uses a dynamic discretization strategy based on Kmeans to make data processing and analysis more accurate and efficient.
5. More comprehensive data feedback: by combining physical and virtual sensor technology, more comprehensive device operation data can be provided, and the device operation data can be better understood by the user.
Drawings
FIG. 1 is a schematic diagram of mining association rules of the present invention;
FIG. 2 is a specific case interface diagram of the mining of association rules of the present invention;
fig. 3 is a specific case interface diagram of the mining result of the association rule of the present invention.
The invention is further illustrated by the following figures and examples, which are not intended to be limiting.
Example 1: the vertical mill energy consumption optimization method based on deep learning and data mining comprises the following steps:
step 1, collecting operation data of vertical mill equipment, and preprocessing the data; in this step, the data acquisition includes:
the temperature sensor is arranged in a critical hot spot area of the vertical mill equipment to monitor the equipment temperature.
The pressure sensor is placed in a significant location where pressure fluctuations may occur.
The rotation speed sensor is used for monitoring the rotation speed of the vertical mill equipment in real time.
Virtual sensor technology is used to estimate key parameters that are difficult to directly measure in conjunction with various physical sensor data. Virtual sensors can provide additional estimation parameters for rule mining, which helps to enhance the accuracy and generalization ability of the model.
The sensor transmits data to the central processing unit in real time through a wireless communication technology for preprocessing.
The data preprocessing can automatically identify and reject abnormal or wrong data points through an abnormal value detection algorithm, a data normalization algorithm ensures that multi-range data are converted into a unified standard, analysis is convenient, and discretization processing is performed on the data after normalization. The discretization processing adopts state cluster analysis; the state clustering analysis is based on a Kmeans algorithm, data to be clustered is selected, each column in the data represents an attribute, each row represents a record of a working condition state, then the number of clusters required by the clustering and the iteration number are set, when the clustering center is not changed or the maximum iteration number is reached, the clustering is stopped, and then the category labeling is completed on the original clustered data. The discretization of data in this step can be specifically seen in table 1-table 5, table 1 is the correspondence between parameters and letter labels (only an outline is given here, specific table contents need to be filled according to actual parameters and labels), table 2-table 4 is the correspondence between partial parameters after discretization and intervals, table 2 is the correspondence table after the feeding amount is discretized, table 3 is the correspondence table after the micro powder ratio table is discretized, table 4 is the correspondence table after the mill shell is vibrated and table 5 is the partial data display after discretization (display of labels of some data points after discretization).
TABLE 1
| Feeding quantity range sign | Value range | Number of elements in range |
| A1 | (0,161.3] | 1423 |
| A2 | (161.3,167.6] | 1268 |
| A3 | (167.6,174.2] | 1847 |
| A4 | (174.2,186] | 853 |
TABLE 2
| Micro powder ratio table range mark | Value range | Number of elements in range |
| B1 | (0,397.3] | 1145 |
| B2 | (397.3,406.6] | 1837 |
| B3 | (406.6,416.7] | 1638 |
| B4 | (416.7,435] | 435 |
TABLE 3 Table 3
TABLE 4 Table 4
TABLE 5
Training the deep learning model by using the data preprocessed in the step 1, and applying the trained model to energy consumption result prediction; and extracting features of the discretized data by using a deep neural network, and automatically learning hidden modes and correlations in the data. This step will generate a feature vector that provides more information for subsequent association rule mining.
The deep learning model adopts a Convolutional Neural Network (CNN) or a long and short time memory network (LSTM) to conduct time sequence analysis of data, and therefore accuracy and effect of association rule mining are improved.
Step 3, performing data mining analysis by using the data preprocessed in the step 1 to acquire association rules; in this step, as shown in fig. 1, the association rule algorithm includes two concepts of support and confidence,
the support is calculated as follows:
the confidence is calculated as follows:
the general association rule algorithm mainly takes the minimum support degree and the minimum confidence degree as constraint conditions, and outputs rules between features meeting the minimum support degree and the minimum confidence degree. In the excavation of the vertical mill energy consumption optimization rule, the aim is to acquire the rule relation between the controllable parameters and the stable index or between the controllable parameters and the energy consumption index of the vertical mill, so that the evaluation function in the association rule algorithm is improved to meet the excavation requirement. The output result of the improved association rule algorithm is a rule between the adjustable parameter and the stability index or between the adjustable parameter and the energy consumption index which meet the requirements of the minimum support degree and the minimum confidence degree. The calculation principle is as follows:
wherein: f (x) is the original evaluation function,the representation rule includes interesting or valid rules.
During operation of the vertical mill, the parameters will fluctuate continuously and not settle at a constant. During control, a parameter is of greater concern than an exact value. The vertical mill data are continuous, so that the invention discretizes the data before the association rule mining, thereby being beneficial to obtaining the association rule which is more in line with the practical application.
And taking the discretized data as input data, and performing association rule mining based on an Apriori algorithm by setting the minimum support and the minimum confidence parameters. As described in fig. 2 and 3, if no valid rules are obtained for a given parameter, or the number of rules is insufficient, it may be necessary to readjust the minimum support and minimum confidence parameters, or to consider further preprocessing of the data.
For example, when the minimum support is set to 6% and the minimum confidence is 75%, 245 association rules are mined from 5390 pieces of data, as shown in tables 6 and 7:
| rules of | support | confidence |
| A1D4H4K4I2 | 0.065 | 1 |
| D4G4I2 | 0.064 | 1 |
| D4H4K4I2 | 0.077 | 0.978 |
| A1D4K4I2 | 0.076 | 0.978 |
| A1D4H4I2 | 0.071 | 0.976 |
| A1H4L4I2 | 0.062 | 0.973 |
| A3F4H4 | 0.075 | 0.974 |
TABLE 6
| Rules of | support | confidence |
| D2K4L2H2 | 0.132 | 0.795 |
| A1D4K4I2L4 | 0.128 | 0.784 |
| A3H2K4L2 | 0.121 | 0.773 |
| A3H2L2K4 | 0.121 | 0.756 |
| C2H2D2I3K4 | 0.14 | 0.954 |
| A1L4K4 | 0.13 | 0.796 |
| A3F4H4 | 0.13 | 0.750 |
TABLE 7
Tables 6 and 7 show examples of rules that are partially mined, e.g., A3F4H4 representation: when the feed amount is in the A3 interval and the mill inlet temperature is in the F4 interval, the vertical mill differential pressure has 75% confidence in the H4 interval.
According to the scheme, by integrating data discretization and Apriori-based rule mining and combining a virtual sensor technology, a more efficient and accurate data preprocessing and analysis method is provided for the vertical mill equipment, and the operation efficiency and the optimal control strategy of the vertical mill equipment are improved.
In a vertical mill plant, virtual sensors are able to estimate certain key parameters, such as parameters related to coal consumption, quality and electricity consumption, by combining data from a plurality of physical sensors. Such newly added information may have an impact on the analysis output of the rules.
1) a1L4K 4-a further analysis of the relationship between feed amount and single ton current using a virtual sensor found that frequent item set correlations of A1L4 were more related to the mass and particle size distribution of the material, which further confirmed the previous conclusions.
2) The correlation of the L2 interval with A3-the virtual sensor may provide deeper information such as humidity or temperature of the material, which may be the reason for the better single ton current at the A3 interval feed.
3) C2H2D2I3K 4-the virtual sensor is used for estimating the actual opening degree of the circulating air valve, so that the heat energy can be better utilized under the condition of the arrangement, and the coal consumption can be reduced. Meanwhile, the practical influence of the A1D4K4I2L4 rule is verified, and the parameter setting is ensured to avoid bad intervals, so that the power consumption is reduced.
4) Design advice-by means of virtual sensors, a more comprehensive running state of the equipment can be obtained, and more accurate and real-time advice is provided. For example, it may be considered to add some virtual sensor captured parameters such as air flow inside the device, the degree of mixing of abrasive with air, etc., which helps to provide more practical and efficient advice to the staff.
5) Rule finishing—in finishing rules, parameters derived from virtual sensors are considered, such as humidity, temperature, or other important parameters of the material. For a rule base for optimizing and regulating energy consumption, the data of the virtual sensor can be considered as a reference to provide more comprehensive and accurate regulation suggestions, and the storage form of the energy consumption rules is shown in table 8:
TABLE 8
The table should include rules derived from the data and virtual sensors, and key indicators such as support and confidence of the rules.
Therefore, by combining the data of the virtual sensor, more comprehensive, accurate and real-time suggestions and strategies can be provided for operators of the vertical mill equipment, so that the operation efficiency of the equipment is effectively improved and the energy consumption is reduced.
And 4, generating an energy consumption optimization strategy by combining the prediction result of the deep learning model in the step 2 and the association rule of the data mining analysis in the step 3, and operating the vertical mill equipment by using the energy consumption optimization strategy.
This step is based on the combined implementation of ARIMA and LSTM models. The ARIMA model performs a corresponding ARIMA operation on the input data and outputs a judgment result. The method comprises the steps that a stationarity detection and a white noise detection output an autocorrelation chart and a partial autocorrelation chart; model scaling outputs the p, d, q values in the model. The prediction module can specify the predicted period number, and in general, the shorter the prediction time is, the more accurate the prediction result of the ARIMA model is. The LSTM model is a deep learning model suitable for time series analysis. In this module, input time series data is normalized first, and then trained and predicted using LSTM. Since LSTM can capture long-term dependencies, it generally performs better than ARIMA on complex time series data. Parameters of the LSTM model, such as the number of layers, the number of neurons, etc., may be adjusted to optimize the predictive performance. After the execution is finished, the user obtains a model report integrating ARIMA and LSTM, wherein the report not only comprises an energy consumption optimization strategy, but also comprises indexes such as coefficients, standard errors, AIC, PIC, p values and the like of the model, and further comprises information such as training loss, verification loss and prediction accuracy of the LSTM model. Wherein the energy consumption optimization strategy may be automatically applied, such as adjusting the rotational speed of the grinding disc, changing the supply of material or adjusting certain parameters in the deep learning model. Or may be semi-automatically applied, i.e., to generate optimization suggestions for operator reference, while the model may be utilized to predictively verify the effectiveness of the suggestions.
Analysis was performed in specific cases: suppose that in "XYZ cement company" the vertical mill equipment is the core part of its main production line. Due to recent increases in energy prices, companies wish to more effectively manage and optimize the energy consumption of their vertical mill equipment. Using the method of the present invention, a company first deploys sensors to collect various operational data of the equipment, such as temperature, pressure, current, material type, and feed rate. These data are input into the self-encoder, resulting in deep features. Based on these deep features, association rule mining reveals the following rules:
when the temperature exceeds 65 ℃, the pressure is lower than 1.2bar, and the material Type is "Type B", a small increase in current is highly correlated to a significant increase in energy consumption.
At currents between 10-12A and feed rates exceeding 5kg/min, the energy consumption of the plant appears to be optimal.
Practical application:
based on the above rules, engineers have formulated the following strategies:
when the apparatus is processing "Type B" material and the temperature is close to 65 ℃, the pressure is monitored and ensured to remain above 1.2 bar.
The current was kept as much as possible in the range of 10-12A and the feed rate was adjusted to not exceed 5kg/min.
After these strategies were implemented, XYZ cement companies observed a reduction in energy consumption of the plant of about 15%.
Conclusion:
by combining deep-learned association rule mining, XYZ cement companies can discover and apply previously unnoticed energy consumption optimization strategies. This not only saves a lot of energy costs for the company, but also improves the production efficiency and the service life of the equipment. This case clearly demonstrates the clear advantage of the process of the invention over the traditional one.
In summary, the invention not only adopts advanced data acquisition and mining algorithm, but also integrates a deep learning model, ensures more efficient and accurate recognition of the energy consumption mode and automatically generates an optimization strategy. The method combining data mining and deep learning overcomes the limitation of the traditional optimization method, and can perform intelligent, accurate and flexible optimization on the energy consumption of the counter grinding equipment. The invention not only can collect the operation condition data of the vertical mill equipment, but also can ensure the accuracy and the real-time performance of the data through a data processing algorithm, and provides solid data support for the operation optimization, the energy consumption monitoring and the fault prediction of the vertical mill equipment.
Claims (9)
1. The vertical mill energy consumption optimization method based on deep learning and data mining is characterized by comprising the following steps of: the method comprises the following steps:
step 1, collecting operation data of vertical mill equipment, and preprocessing the data;
training the deep learning model by using the data preprocessed in the step 1, and applying the trained model to energy consumption result prediction;
step 3, performing data mining analysis by using the data preprocessed in the step 1 to acquire association rules;
and 4, generating an energy consumption optimization strategy by combining the prediction result of the deep learning model in the step 2 and the association rule of the data mining analysis in the step 3, and operating the vertical mill equipment by using the energy consumption optimization strategy.
2. The vertical mill energy consumption optimization method based on deep learning and data mining according to claim 1, wherein the method comprises the following steps: in the step 1, preprocessing of data comprises the steps of removing abnormal values and noise by using a data cleaning method, then normalizing and normalizing the data, and then discretizing the data.
3. The vertical mill energy consumption optimization method based on deep learning and data mining according to claim 2, wherein the method is characterized in that: the discretization processing adopts state cluster analysis; the state clustering analysis is based on a Kmeans algorithm, data to be clustered is selected, each column in the data represents an attribute, each row represents a record of a working condition state, then the number of clusters required by the clustering and the iteration number are set, when the clustering center is not changed or the maximum iteration number is reached, the clustering is stopped, and then the category labeling is completed on the original clustered data.
4. The vertical mill energy consumption optimization method based on deep learning and data mining according to claim 1, wherein the method comprises the following steps: in step 2, the deep learning model is a convolutional neural network, a cyclic neural network and/or a long-short-term memory network.
5. The vertical mill energy consumption optimization method based on deep learning and data mining according to claim 2, wherein the method is characterized in that: in the step 3, discretized data is used as input data, and association rule mining is performed based on an Apriori algorithm by setting minimum support and minimum confidence parameters.
6. The vertical mill energy consumption optimization method based on deep learning and data mining according to claim 5, wherein the method is characterized in that: the support is calculated as follows:
the confidence is calculated as follows:
7. the vertical mill energy consumption optimization method based on deep learning and data mining according to claim 5, wherein the method is characterized in that: the output result of the evaluation function of the Apriori algorithm is a rule between an adjustable parameter and a stable index or between an adjustable parameter and an energy consumption index meeting the requirements of minimum support and minimum confidence coefficient, and the rule is calculated as follows:
wherein: f (x) is the original evaluation function,the representation rule includes interesting or valid rules.
8. The system of the vertical mill energy consumption optimization method based on deep learning and data mining according to claim 1, wherein: the system comprises a data acquisition module, a data preprocessing module, a data mining module and an optimization strategy generation module;
the data acquisition module acquires data of the vertical face equipment by using various sensors;
the data preprocessing module applies an algorithm to perform data preprocessing;
the data mining module uses a data mining algorithm to analyze real-time data;
and the optimization strategy generation module generates energy consumption optimization strategies and rules according to the data mining result.
9. The system of the vertical mill energy consumption optimization method based on deep learning and data mining according to claim 8, wherein: the data acquisition module also comprises a virtual sensor module which is used for simulating and predicting certain equipment parameters which cannot be directly measured through an algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410045317.0A CN117874464A (en) | 2024-01-12 | 2024-01-12 | Vertical mill energy consumption optimization method based on deep learning and data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410045317.0A CN117874464A (en) | 2024-01-12 | 2024-01-12 | Vertical mill energy consumption optimization method based on deep learning and data mining |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117874464A true CN117874464A (en) | 2024-04-12 |
Family
ID=90589822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410045317.0A Pending CN117874464A (en) | 2024-01-12 | 2024-01-12 | Vertical mill energy consumption optimization method based on deep learning and data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117874464A (en) |
-
2024
- 2024-01-12 CN CN202410045317.0A patent/CN117874464A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104808587B (en) | A kind of mobility statistical method based on machining apparatus running status | |
| CN110503338A (en) | A kind of ubiquitous electric power Internet of Things monitoring method | |
| CN102486833B (en) | Method for predicting efficiency and detecting fault of device | |
| CN114723285B (en) | Power grid equipment safety evaluation prediction method | |
| CN103631681A (en) | Method for online restoring abnormal data of wind power plant | |
| CN108053128A (en) | A kind of Power Network Transient Stability fast evaluation method based on ELM and TF | |
| CN107291830A (en) | A kind of creation method of equipment knowledge base | |
| CN111628494B (en) | Low-voltage distribution network topology identification method and system based on logistic regression method | |
| CN112926749A (en) | Intelligent power grid information equipment monitoring system and method | |
| CN117214637B (en) | Charging gun withstand voltage test method, device, equipment and storage medium | |
| CN114113773A (en) | Non-invasive load monitoring method based on zero sample learning | |
| CN117878979B (en) | Power balance and dynamic compensation system based on electric energy storage | |
| CN117469603B (en) | Multi-water-plant water supply system pressure optimal control method based on big data learning | |
| CN113722656B (en) | Real-time health evaluation method and system for thermal generator set | |
| CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
| CN117874464A (en) | Vertical mill energy consumption optimization method based on deep learning and data mining | |
| KR102126954B1 (en) | Method for measuring smart sewing works and system for performing the same | |
| CN115829122A (en) | Electric power customer service work order complaint early warning method and system | |
| CN115130805A (en) | Analysis device, analysis method, and computer-readable medium having program recorded thereon | |
| CN110659681A (en) | Time sequence data prediction system and method based on pattern recognition | |
| CN117054734B (en) | Method and system for monitoring smelting energy consumption of copper plate | |
| CN117094478B (en) | Energy scheduling management method, device, equipment and storage medium | |
| CN113591283B (en) | Method and device for adjusting operating oxygen amount of gas boiler and computer equipment | |
| CN111682574B (en) | Method for identifying running scene of alternating current-direct current hybrid system, storage medium and equipment | |
| CN111752791A (en) | Intelligent adjusting system and method for computer operation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |