CN115294671A - Air compressor outlet pressure prediction method and prediction system - Google Patents
Air compressor outlet pressure prediction method and prediction system Download PDFInfo
- Publication number
- CN115294671A CN115294671A CN202210944823.4A CN202210944823A CN115294671A CN 115294671 A CN115294671 A CN 115294671A CN 202210944823 A CN202210944823 A CN 202210944823A CN 115294671 A CN115294671 A CN 115294671A
- Authority
- CN
- China
- Prior art keywords
- air compressor
- outlet pressure
- data
- network structure
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 25
- 230000002159 abnormal effect Effects 0.000 claims abstract description 20
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000010219 correlation analysis Methods 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 16
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 238000012417 linear regression Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000004590 computer program Methods 0.000 claims 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000010220 Pearson correlation analysis Methods 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000011425 standardization method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/005—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles during manufacturing process
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses a prediction method of air compressor outlet pressure, which relates to the field of data processing and comprises the following steps: acquiring running data of an air compressor, removing abnormal data and carrying out standardized processing; performing correlation analysis on the operation data of the air compressor, and selecting operation parameters related to the outlet pressure of the air compressor as network input; training a multilayer perceptron network structure by taking operation parameters related to the outlet pressure of the air compressor as sample data, selecting the operation parameters as feature sets to train the multilayer perceptron network structure, optimizing the multilayer perceptron network structure by using an Adam optimization algorithm, and minimizing a loss function; and obtaining a prediction model, and predicting the outlet pressure of the air compressor by using the prediction model.
Description
Technical Field
The application relates to the field of data processing, in particular to a prediction method and a prediction system for outlet pressure of an air compressor.
Background
For the air compressor system, the prediction of parameters by using a data mining technology has great significance. Data mining techniques can be divided into prediction models, classification and regression, cluster analysis, association analysis, sequence pattern discovery, dependency models, fault and outlier detection, and the like according to mining tasks. The artificial neural network technology is used as a mature theory and is more and more widely applied to simulation prediction, and the application range of the artificial neural network technology can cover the aspects of production control, process design, production scheduling and the like.
Meanwhile, with the rapid development of the intellectualization of industrial production, the monitoring of the real-time operation state of mechanical equipment, the processing of abnormal faults of the equipment and the prediction of the operation parameters of the equipment with high precision become the key research direction in the current industrial industry. The air compressor is one of key equipment in the air separation plant, provides compressed air with specified pressure for a subsequent air cooling tower and a refrigeration machine, and provides required air raw materials for the rectifying tower. Because the state of the compressor changes at any time, the outlet pressure of the air compressor directly influences the actual shaft power and the safety, efficiency, energy consumption and the like of the whole air compressor system. Therefore, the research on the prediction algorithm of the outlet pressure of the air compressor has important significance for researching the safety of the air compressor, the production efficiency of a factory, the energy conservation and the like. In the traditional data prediction monitoring process, a prediction model is established by adopting a machine learning algorithm, so that data is predicted. However, the analysis of the data of the air compressor equipment parameters is very complicated, the precision of the prediction result of a common prediction model is not enough, and the error is large.
Disclosure of Invention
The method is a novel method for predicting the outlet pressure of the air compressor, and is provided for monitoring the outlet pressure of the air compressor aiming at the problems that the outlet pressure of the air compressor in a factory is abnormal in the running process and belongs to common faults and the characteristics of nonlinearity, atypical property and nonequivalence of parameters of the air compressor.
In order to achieve the purpose, the following technical scheme is adopted in the application:
a method for predicting the outlet pressure of an air compressor comprises the following steps:
acquiring the operation data of the air compressor, eliminating abnormal data and carrying out standardized processing;
performing correlation analysis on the operation data of the air compressor, and selecting operation parameters related to the outlet pressure of the air compressor as network input;
training a multi-layer perceptron network structure by taking operation parameters related to the outlet pressure of the air compressor as sample data, selecting the operation parameters as feature sets to train the multi-layer perceptron network structure, optimizing the multi-layer perceptron network structure by using an Adam optimization algorithm, and minimizing a loss function;
and obtaining a prediction model, and predicting the outlet pressure of the air compressor by using the prediction model.
Preferably, the method for eliminating abnormal data and performing normalization processing includes:
the method comprises a plurality of parameter types, and abnormal data is removed from sample data of each parameter type according to a 3 sigma criterion.
Preferably, the method for performing correlation analysis on the air compressor operation data and selecting the operation parameters related to the outlet pressure of the air compressor as network input specifically comprises the following steps:
and performing correlation analysis by adopting a Pearson correlation coefficient, taking a parameter matrix of the main parameter type as an input variable, taking an outlet pressure value of the air compressor as an output variable, and selecting data corresponding to the main parameter type as network input according to the correlation coefficient.
Preferably, the method for optimizing the network structure of the multi-layer perceptron by using the Adam optimization algorithm and minimizing the loss function specifically includes:
and the output layer outputs the predicted outlet pressure of the air compressor, the predicted outlet pressure of the air compressor is given to a loss function used in the linear regression, and the optimization target is a weight parameter and an offset parameter.
Preferably, the loss function of the prediction model adopts a root mean square error RMSE, which specifically includes:
predicting results with a model
And comparing the data with actual air compressor data, wherein the calculation formula is as follows:
and N is a data volume, and the model precision is adjusted by analyzing the value of RMSE to obtain an optimal model.
Preferably, the feature set comprises a training set and a test set, the training set is used for training a network structure of the multilayer perceptron, the test set test data is used for forecasting, and the actual outlet pressure of the air compressor is compared with the forecast outlet pressure of the air compressor.
Preferably, a multi-layer perceptron network structure with double hidden layers is adopted, and hidden nodes of the first hidden layer are larger than those of the second hidden layer.
Based on the air compressor outlet pressure prediction method, the air compressor outlet pressure prediction system is further disclosed, and the air compressor outlet pressure prediction system comprises the following structures:
the preprocessing unit is used for acquiring the running data of the air compressor, eliminating abnormal data and carrying out standardized processing;
the correlation analysis unit is used for carrying out correlation analysis on the operation data of the air compressor and selecting operation parameters related to the outlet pressure of the air compressor as network input;
the model training unit is used for training the network structure of the multilayer perceptron by taking the operating parameters related to the outlet pressure of the air compressor as sample data, selecting the operating parameters as characteristic sets to train the network structure of the multilayer perceptron, optimizing the network structure of the multilayer perceptron by an Adam optimization algorithm, and minimizing a loss function to obtain an optimal model;
and the model test unit is used for calculating the outlet pressure of the air compressor through test collection and measurement.
The input unit inputs the running data of the air compressor;
and the output unit is used for outputting the predicted outlet pressure of the air compressor.
The invention has the following beneficial effects:
(1) The multi-layer perceptron adopted by the invention is a feedforward artificial neural network model, and a plurality of input data sets are mapped to a single output data set, so that the problem of any linear inseparability is solved, the health analysis of the air compressor system is effectively realized, and the multi-layer perceptron can be further combined with a digital twin system to guide the field operation, thereby ensuring the safe and efficient operation of field equipment.
(2) Because the internal operation parameters of the air compressor are very complex, in order to reduce the training difficulty and improve the training efficiency, some parameters with low correlation with the network model need to be removed during data processing, and the parameter data with high correlation with the network model is mainly researched.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the description below are only some embodiments of the present application, and for those skilled in the art, other drawings may be obtained according to these drawings without inventive labor.
FIG. 1 is a flow chart of a method for predicting the outlet pressure of an air compressor;
FIG. 2 is a schematic structural diagram of a double hidden layer MLP model;
FIG. 3 is a parameter diagram of a double hidden layer MLP model;
fig. 4 is a comparison graph of the predicted value and the actual data of the outlet pressure of the air compressor.
Detailed Description
The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first," "second," and the like in the claims and in the description of the application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order, it being understood that the terms so used are interchangeable under appropriate circumstances and are merely used to describe a different manner of distinguishing between similar elements in the embodiments of the application and that the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion such that a process, method, system, article, or apparatus that comprises a list of elements is not necessarily limited to those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1:
a method for predicting the outlet pressure of an air compressor comprises the following steps:
(1) Acquiring the running data of the air compressor, eliminating abnormal data and carrying out standardized processing;
(2) Performing correlation analysis on the operation data of the air compressor, and selecting operation parameters related to the outlet pressure of the air compressor as network input;
(3) Training a multi-layer perceptron network structure by taking operation parameters related to the outlet pressure of the air compressor as sample data, selecting the operation parameters as feature sets to train the multi-layer perceptron network structure, optimizing the multi-layer perceptron network structure by using an Adam optimization algorithm, and minimizing a loss function;
(4) And training to obtain a prediction model so as to predict the outlet pressure of the air compressor by using the prediction model.
In step (1), the method for eliminating abnormal data and performing standardization processing comprises the following steps: extracting a plurality of parameter types, and eliminating abnormal data from the sample data of each parameter type according to a 3 sigma criterion.
Let each type of parameter sample data be x 1 ，x 2 ，x 3 ，…x N Average value of
Deviation is as
Calculating the standard deviation according to Bessel formula
If a certain sample data x i Deviation v of i Satisfy | v i If the data is considered abnormal data, the data is considered to be removed if the data is greater than 3 sigma.
Abnormal value analysis can check whether the system uploads data with entry errors or not and whether the data contain abnormal data. The abnormal values are put into the calculation and analysis process of the data without removing the abnormal values, and the prediction result of the outlet pressure of the air compressor is adversely affected. For the standardization processing method, a 0-1 standardization method is adopted, so that dimension influence among indexes can be eliminated, and comparability among data indexes is solved.
And (3) performing correlation analysis on the operating data of the air compressor, selecting operating parameters related to the outlet pressure of the air compressor as network input, performing correlation analysis by adopting a Pearson correlation coefficient, selecting a main parameter matrix as an input variable, selecting an outlet pressure value of the air compressor as an output variable, and selecting main correlation parameters as network input according to the correlation coefficient.
Specifically, the formula of the Pearson correlation coefficient is as follows:
where X (X parameter matrices) and Y (outlet pressure) are input and output variables, respectively, and N is the magnitude of the observed quantity (e.g., 1000 samples). And (3) collecting the parameter data of the normal operation of each device in the air compressor on the spot, and calculating to obtain the correlation coefficient. The correlation coefficient is specified to be very strong in the range of 0.8 to 1.0, strong in the range of 0.6 to 0.8, moderate in the range of 0.4 to 0.6, and almost irrelevant in the range of 0 to 0.4. And selecting parameters with extremely strong correlation coefficients as input values.
Because the air compressor device has thousands of measuring points, but not every measuring point has a relationship with the outlet pressure of the air compressor, wherein only part of main parameters which have significant influence on the outlet pressure of the air compressor can be determined, and according to the above Pearson correlation analysis, the following parameters can be determined as the main parameters: air compressor inlet temperature, air compressor inlet pressure, air compressor return water pressure, air compressor inlet water temperature, air compressor return water temperature, air compressor outlet flow rate, air compressor instantaneous flow, air compressor outlet temperature, air compressor circulating water inlet pressure, air compressor circulating water return water pressure, air compressor circulating water inlet temperature, air compressor circulating water return water temperature, air compressor tertiary outlet pressure, air compressor current, air compressor voltage, and air compressor power. These 17 parameters are used as network inputs.
Step (3), as shown in fig. 2 and fig. 3, a deep learning model is built, a multi-layer perceptron network structure is selected, an input layer, a hidden layer and an output layer are configured, in the embodiment, double hidden layers are adopted, the size of the hidden layer is 20,12, specifically, after the processing of the steps (1) and (2), a given sample X belongs to R n×d The batch size is n, and the input number is d. In the embodiment, d is the number of input variables 17, n is the length of each input variable during model training, and adjustment is performed according to data stored in the air compressor data acquisition system. A hidden layer of the multi-layer perceptron is assumed, wherein the number of hidden units is h. The output of the hidden layer (also called hidden layer variable or hidden variable) is recorded as H, and H is equal to R n×h . Because the hidden layer and the output layer are all connected layers, the weight parameter and the bias parameter of the hidden layer can be set as W respectively h ∈R d×h And b h ∈R 1×h The weight and deviation parameters of the output layer are W o ∈R h×q And b o ∈R 1×q . The multilayer perceptron hidden layer calculates the output in the following way:
H＝φ(XW h +b h )
O＝HW o +b o (2)
where φ represents an activation function and the expression is
The function of the activation function is to introduce nonlinear operation into the learning network, so that the network can approach any nonlinear function, and the generalization capability of the model is greatly improved. In the invention, a double-layer hidden layer is selected, more hidden nodes are arranged on the 1 st hidden layer, and less hidden nodes are arranged on the 2 nd hidden layer, so that the performance of a multi-layer feedforward network is improved, and the prediction effect is improved. The network structure is thus (17, 20,12, 1). The parameter at the output level is the predicted compressor outlet pressure.
Further, a multi-layer perceptron network structure is optimized by an Adam optimization algorithm, and parameters of an output layer are directly provided for a square loss function used in linear regression. The optimization target is a weight parameter and an offset parameter, so that the loss function is minimum, which means that the predicted outlet pressure of the air compressor output by the network is closer to the actual outlet pressure of the air compressor.
Adam is adopted in the optimization method, firstly, the optimization method is simple to implement, high in calculation efficiency and capable of saving analysis time; secondly, a large amount of historical data exist in the air compressor data acquisition system, a large amount of data are often selected for model training in consideration of prediction accuracy, and the optimization algorithm is more suitable for the operation scene; in addition, the air compressor running data acquired by the sensor cannot be influenced by noise, and the influence of the noise on the model precision can be reduced by the optimization algorithm.
The loss function of the prediction model uses the root mean square error RMSE, where the model predicts the result
And comparing the data with actual air compressor data, wherein the calculation formula is as follows, and N is the data volume:
and minimizing the root mean square error RMSE, if so, saving the model and entering the next test set for verification, otherwise, adjusting the weight parameter and the bias parameter to continue training, and circularly judging until the prediction model is finished.
The sample data comprises a training set and a testing set, the training set is adopted by a training model, 1000 sample data are selected in the embodiment at the testing stage, and the predicted outlet pressure of the air compressor is calculated according to model parameters stored in the testing stage. Meanwhile, before the model training is started, the hyper-parameters in the network are set in advance, and generally need to be set through cross validation.
In summary, based on relevant factors affecting the outlet pressure of the air compressor, the obtained real data is combined, the outlet pressure is predicted by using the MLP algorithm, and an attempt is made for further development in the field.
As can be seen from fig. 4, the predicted values differ little from the true values, RMSE =0.001. I.e. the network prediction results meet the requirements. The air compressor outlet pressure prediction algorithm based on the multilayer perceptron has good fitting capacity and is worthy of being applied to practice.
Example 2:
based on the air compressor outlet pressure prediction method, the air compressor outlet pressure prediction system is further disclosed, and comprises the following structures:
the preprocessing unit is used for acquiring the running data of the air compressor, eliminating abnormal data and carrying out standardized processing;
the correlation analysis unit is used for performing correlation analysis on the operating data of the air compressor and selecting operating parameters related to the outlet pressure of the air compressor as network input;
the model training unit is used for training the network structure of the multilayer perceptron by taking the operating parameters related to the outlet pressure of the air compressor as sample data, selecting the operating parameters as characteristic sets to train the network structure of the multilayer perceptron, optimizing the network structure of the multilayer perceptron by an Adam optimization algorithm, and minimizing a loss function to obtain an optimal model;
and the model test unit is used for calculating the outlet pressure of the air compressor through test collection and measurement.
The input unit is used for inputting the operation data of the air compressor;
and the output unit is used for outputting the predicted outlet pressure of the air compressor.
In the several embodiments disclosed in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the modules or units is only one logical functional division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another device, or some features may be omitted, or not executed.
The units may or may not be physically separate, and components displayed as units may be one physical unit or a plurality of physical units, that is, may be located in one place, or may be distributed in a plurality of different places. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a readable storage medium. Based on such understanding, the technical solution of the embodiments of the present invention may essentially or partially contribute to the prior art, or all or part of the technical solution may be embodied in the form of a software product, where the software product is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, etc.) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. The method for predicting the outlet pressure of the air compressor is characterized by comprising the following steps of:
acquiring the operation data of the air compressor, eliminating abnormal data and carrying out standardized processing;
performing correlation analysis on the operation data of the air compressor, and selecting operation parameters related to the outlet pressure of the air compressor as network input;
training a multilayer perceptron network structure by taking operation parameters related to the outlet pressure of the air compressor as sample data, selecting the operation parameters as feature sets to train the multilayer perceptron network structure, optimizing the multilayer perceptron network structure by using an Adam optimization algorithm, and minimizing a loss function;
and obtaining a prediction model, and predicting the outlet pressure of the air compressor by using the prediction model.
2. The air compressor outlet pressure prediction method according to claim 1, wherein the method for eliminating abnormal data and performing normalization processing comprises the following steps:
the method comprises a plurality of parameter types, and abnormal data is removed from sample data of each parameter type according to a 3 sigma criterion.
3. The method for predicting the outlet pressure of the air compressor as claimed in claim 1, wherein the method for performing correlation analysis on the operating data of the air compressor and selecting the operating parameters related to the outlet pressure of the air compressor as network inputs specifically comprises:
and performing correlation analysis by adopting a Pearson correlation coefficient, taking a parameter matrix of a main parameter type as an input variable, taking an outlet pressure value of the air compressor as an output variable, and selecting data corresponding to the main parameter type according to the correlation coefficient as network input.
4. The air compressor outlet pressure prediction method according to claim 1, wherein the method for optimizing the network structure of the multilayer perceptron and minimizing the loss function by using an Adam optimization algorithm specifically comprises the following steps:
and the output layer outputs the predicted outlet pressure of the air compressor, the predicted outlet pressure of the air compressor is given to a loss function used in the linear regression, and the optimization target is a weight parameter and an offset parameter.
5. The air compressor outlet pressure prediction method according to claim 1 or 4, wherein a loss function of the prediction model adopts a Root Mean Square Error (RMSE), and specifically comprises:
predicting results with a model
And comparing the data with actual air compressor data, wherein the calculation formula is as follows:
and N is a data volume, and the model precision is adjusted by analyzing the value of RMSE to obtain an optimal model.
6. The method of claim 1, wherein the feature set comprises a training set and a test set, the training set is used to train a multi-level sensor network structure, the test set test data is used to predict, and actual compressor outlet pressure is compared with predicted compressor outlet pressure.
7. The method for predicting the outlet pressure of the air compressor as claimed in claim 1, wherein a multi-layer sensor network structure with double hidden layers is adopted, and hidden nodes of a first hidden layer are larger than those of a second hidden layer.
8. The utility model provides an air compressor machine outlet pressure prediction system which characterized in that includes following structure:
the preprocessing unit is used for acquiring the running data of the air compressor, eliminating abnormal data and carrying out standardized processing;
the correlation analysis unit is used for performing correlation analysis on the operating data of the air compressor and selecting operating parameters related to the outlet pressure of the air compressor as network input;
the model training unit is used for training the network structure of the multilayer perceptron by taking the operating parameters related to the outlet pressure of the air compressor as sample data, selecting the operating parameters as characteristic sets to train the network structure of the multilayer perceptron, optimizing the network structure of the multilayer perceptron by an Adam optimization algorithm, and minimizing a loss function to obtain an optimal model;
the model test unit is used for calculating the outlet pressure of the air compressor by test set measurement;
the input unit is used for inputting the operation data of the air compressor;
and the output unit is used for outputting the predicted outlet pressure of the air compressor.
9. A computer device, characterized in that the computer device comprises a memory and a processor, the memory stores a computer program, the processor when executing the computer program realizes the air compressor outlet pressure prediction method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that the storage medium stores a computer program which, when being executed by a processor, can implement the air compressor outlet pressure prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210944823.4A CN115294671A (en) | 2022-08-08 | 2022-08-08 | Air compressor outlet pressure prediction method and prediction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210944823.4A CN115294671A (en) | 2022-08-08 | 2022-08-08 | Air compressor outlet pressure prediction method and prediction system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115294671A true CN115294671A (en) | 2022-11-04 |
Family
ID=83828109
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210944823.4A Pending CN115294671A (en) | 2022-08-08 | 2022-08-08 | Air compressor outlet pressure prediction method and prediction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115294671A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117032120A (en) * | 2023-08-16 | 2023-11-10 | 广东超级龙节能环保科技有限公司 | Integrated intelligent cloud control system and control method for air compression station |
| CN117236511A (en) * | 2023-09-26 | 2023-12-15 | 中交广州航道局有限公司 | Big data prediction method and device for vacuum degree of underwater pump of cutter suction dredger |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060184255A1 (en) * | 2005-02-11 | 2006-08-17 | Roger Dixon | Adaptive sensor model |
| US20110067396A1 (en) * | 2009-09-22 | 2011-03-24 | Gm Global Technology Operations, Inc. | Pressure estimation systems and methods |
| CN102831475A (en) * | 2012-08-01 | 2012-12-19 | 华北电力大学 | Sine normalization method for power forecast model of wind power plant |
| TW201310260A (en) * | 2011-08-16 | 2013-03-01 | Advanced Control & Systems Inc | Forecasting management method for air compressor operation |
| CN103745271A (en) * | 2014-01-02 | 2014-04-23 | 上海大学 | Forecasting method for induction thermal deposition calcium-phosphate coating process on basis of neural network |
| CN104504297A (en) * | 2015-01-21 | 2015-04-08 | 甘肃百合物联科技信息有限公司 | Method for using neural network to forecast hypertension |
| CN110503133A (en) * | 2019-07-26 | 2019-11-26 | 东北大学 | A kind of centrifugal compressor failure prediction method based on deep learning |
| CN111178978A (en) * | 2019-12-31 | 2020-05-19 | 中山大学 | Air ticket price prediction method combining flight information and price sequence |
| CN111950854A (en) * | 2020-07-14 | 2020-11-17 | 包头钢铁（集团）有限责任公司 | Coke quality index prediction method based on multilayer neural network |
| CN111970163A (en) * | 2020-06-30 | 2020-11-20 | 网络通信与安全紫金山实验室 | Network flow prediction method of LSTM model based on attention mechanism |
| CN112488395A (en) * | 2020-12-01 | 2021-03-12 | 湖南大学 | Power distribution network line loss prediction method and system |
| CN112529320A (en) * | 2020-12-18 | 2021-03-19 | 上海应用技术大学 | Intelligent diagnosis system for air compressor cluster |
| CN113339310A (en) * | 2021-07-19 | 2021-09-03 | 东北大学 | Compressor surge prediction control method based on magnetic suspension axial position control |
| CN113554466A (en) * | 2021-07-26 | 2021-10-26 | 国网四川省电力公司电力科学研究院 | Short-term power consumption prediction model construction method, prediction method and device |
| CN114169645A (en) * | 2021-12-31 | 2022-03-11 | 中国南方电网有限责任公司 | Short-term load prediction method for smart power grid |
| CN114182510A (en) * | 2021-12-31 | 2022-03-15 | 广东美的白色家电技术创新中心有限公司 | Clothes dryer control method and device, clothes dryer and storage medium |
| CN114295556A (en) * | 2021-12-28 | 2022-04-08 | 浙江工业大学 | Hyperspectral physicochemical index fitting method and system based on one-dimensional convolution |
| CN114492163A (en) * | 2021-12-24 | 2022-05-13 | 武汉钢铁有限公司 | Blast furnace fan operation condition prediction method, device, equipment and storage medium |
| CN114547974A (en) * | 2022-02-11 | 2022-05-27 | 齐鲁工业大学 | Dynamic soft measurement modeling method based on input variable selection and LSTM neural network |
-
2022
- 2022-08-08 CN CN202210944823.4A patent/CN115294671A/en active Pending
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060184255A1 (en) * | 2005-02-11 | 2006-08-17 | Roger Dixon | Adaptive sensor model |
| US20110067396A1 (en) * | 2009-09-22 | 2011-03-24 | Gm Global Technology Operations, Inc. | Pressure estimation systems and methods |
| TW201310260A (en) * | 2011-08-16 | 2013-03-01 | Advanced Control & Systems Inc | Forecasting management method for air compressor operation |
| CN102831475A (en) * | 2012-08-01 | 2012-12-19 | 华北电力大学 | Sine normalization method for power forecast model of wind power plant |
| CN103745271A (en) * | 2014-01-02 | 2014-04-23 | 上海大学 | Forecasting method for induction thermal deposition calcium-phosphate coating process on basis of neural network |
| CN104504297A (en) * | 2015-01-21 | 2015-04-08 | 甘肃百合物联科技信息有限公司 | Method for using neural network to forecast hypertension |
| CN110503133A (en) * | 2019-07-26 | 2019-11-26 | 东北大学 | A kind of centrifugal compressor failure prediction method based on deep learning |
| CN111178978A (en) * | 2019-12-31 | 2020-05-19 | 中山大学 | Air ticket price prediction method combining flight information and price sequence |
| CN111970163A (en) * | 2020-06-30 | 2020-11-20 | 网络通信与安全紫金山实验室 | Network flow prediction method of LSTM model based on attention mechanism |
| CN111950854A (en) * | 2020-07-14 | 2020-11-17 | 包头钢铁（集团）有限责任公司 | Coke quality index prediction method based on multilayer neural network |
| CN112488395A (en) * | 2020-12-01 | 2021-03-12 | 湖南大学 | Power distribution network line loss prediction method and system |
| CN112529320A (en) * | 2020-12-18 | 2021-03-19 | 上海应用技术大学 | Intelligent diagnosis system for air compressor cluster |
| CN113339310A (en) * | 2021-07-19 | 2021-09-03 | 东北大学 | Compressor surge prediction control method based on magnetic suspension axial position control |
| CN113554466A (en) * | 2021-07-26 | 2021-10-26 | 国网四川省电力公司电力科学研究院 | Short-term power consumption prediction model construction method, prediction method and device |
| CN114492163A (en) * | 2021-12-24 | 2022-05-13 | 武汉钢铁有限公司 | Blast furnace fan operation condition prediction method, device, equipment and storage medium |
| CN114295556A (en) * | 2021-12-28 | 2022-04-08 | 浙江工业大学 | Hyperspectral physicochemical index fitting method and system based on one-dimensional convolution |
| CN114169645A (en) * | 2021-12-31 | 2022-03-11 | 中国南方电网有限责任公司 | Short-term load prediction method for smart power grid |
| CN114182510A (en) * | 2021-12-31 | 2022-03-15 | 广东美的白色家电技术创新中心有限公司 | Clothes dryer control method and device, clothes dryer and storage medium |
| CN114547974A (en) * | 2022-02-11 | 2022-05-27 | 齐鲁工业大学 | Dynamic soft measurement modeling method based on input variable selection and LSTM neural network |
Non-Patent Citations (1)
| Title |
|---|
| 刘树春 等: "《深度实践OCR 基于深度学习的文字识别》", vol. 1, 31 May 2020, 机械工业出版社, pages: 154 - 156 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117032120A (en) * | 2023-08-16 | 2023-11-10 | 广东超级龙节能环保科技有限公司 | Integrated intelligent cloud control system and control method for air compression station |
| CN117236511A (en) * | 2023-09-26 | 2023-12-15 | 中交广州航道局有限公司 | Big data prediction method and device for vacuum degree of underwater pump of cutter suction dredger |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fu et al. | Condition monitoring of wind turbine gearbox bearing based on deep learning model | |
| CN115294671A (en) | Air compressor outlet pressure prediction method and prediction system | |
| CN111539553B (en) | Wind turbine generator fault early warning method based on SVR algorithm and off-peak degree | |
| US11840998B2 (en) | Hydraulic turbine cavitation acoustic signal identification method based on big data machine learning | |
| CN112101480A (en) | Multivariate clustering and fused time sequence combined prediction method | |
| CN110262450B (en) | Fault prediction method for cooperative analysis of multiple fault characteristics of steam turbine | |
| CN113107626B (en) | Load prediction method of combined cycle generator set based on multivariable LSTM | |
| CN111525587B (en) | Reactive load situation-based power grid reactive voltage control method and system | |
| CN112132394B (en) | Power plant circulating water pump predictive state evaluation method and system | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN115238573A (en) | Hydroelectric generating set performance degradation trend prediction method and system considering working condition parameters | |
| CN114757269A (en) | Complex process refined fault detection method based on local subspace-neighborhood preserving embedding | |
| CN114218872A (en) | Method for predicting remaining service life based on DBN-LSTM semi-supervised joint model | |
| CN115313930A (en) | Method for predicting excitation current of large-scale steam turbine generator in real time | |
| CN114429238A (en) | Wind turbine generator fault early warning method based on space-time feature extraction | |
| CN116127831A (en) | Soft measurement method for difficult-to-measure parameters of heavy gas turbine | |
| Lyu et al. | Interval Prediction of Remaining Useful Life based on Convolutional Auto-Encode and Lower Upper Bound Estimation | |
| CN115496264A (en) | Method for predicting generated power of wind turbine generator | |
| CN112628132B (en) | Water pump key index prediction method based on machine learning | |
| CN115578084A (en) | Wind turbine generator set frequency converter fault early warning method based on deep convolution self-encoder | |
| CN115238854A (en) | Short-term load prediction method based on TCN-LSTM-AM | |
| CN114298413A (en) | Hydroelectric generating set runout trend prediction method | |
| Sanayha et al. | Remaining useful life prediction using enhanced convolutional neural network on multivariate time series sensor data | |
| Ou et al. | Fault Prediction Model of Wind Power Pitch System Based on BP Neural Network | |
| Xu et al. | A method for predicting the remaining life of equipment based on WTTE-CNN-LSTM |
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 |