CN114004391A - Power equipment extreme condition performance prediction system based on Gaussian process regression - Google Patents

Power equipment extreme condition performance prediction system based on Gaussian process regression Download PDF

Info

Publication number
CN114004391A
CN114004391A CN202111201676.3A CN202111201676A CN114004391A CN 114004391 A CN114004391 A CN 114004391A CN 202111201676 A CN202111201676 A CN 202111201676A CN 114004391 A CN114004391 A CN 114004391A
Authority
CN
China
Prior art keywords
data
gaussian process
process regression
model
training
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
Application number
CN202111201676.3A
Other languages
Chinese (zh)
Inventor
陈俐
王子垚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202111201676.3A priority Critical patent/CN114004391A/en
Publication of CN114004391A publication Critical patent/CN114004391A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • 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 relates to a power equipment extreme condition performance prediction system based on Gaussian process regression, which comprises: the conventional working condition data acquisition unit: the system is used for dynamically acquiring system input, system output and environmental data of the power equipment in an experimental environment through the sensing equipment, namely conventional working condition data; a terminal processing unit: the data transmission unit is used for receiving the data of the conventional working conditions and sending the data to the server unit through the data transmission unit; a server unit: and constructing a Gaussian process regression model, training standard working conditions and extreme working conditions to obtain a trained Gaussian process regression model, feeding back the trained Gaussian process regression model to a terminal processing unit through a data transmission unit, manually inputting data through a human-computer interaction unit, predicting the performance of the extreme working conditions of the power equipment, and displaying the prediction result through the human-computer interaction unit. Compared with the prior art, the method has the advantages of performance prediction under all working conditions, experiment cost reduction and the like.

Description

Power equipment extreme condition performance prediction system based on Gaussian process regression
Technical Field
The invention relates to the field of detection of working conditions of power equipment, in particular to a performance prediction system for extreme working conditions of power equipment based on Gaussian process regression and generation countermeasure network.
Background
In the field of industrial technology, prediction of power system performance is particularly important for evaluating the quality of a power system. The existing methods for evaluating the advantages and the disadvantages of the power system have two types, one is to carry out a physical experiment and detect the performance of the power system in an experimental environment; and the other method is to establish a physical model and a computer simulation model of the power system, further perform a simulation experiment of the power system and detect the performance of the power system. The former needs to build a relatively perfect experiment table, consumes a large amount of manpower, material resources and time resources, and the physical experiment can simulate limited working conditions, so that the performance of the power system under some extreme working conditions cannot be detected; in the latter, because the action mechanism of the power system is complex, various factors influencing the performance of the power system are difficult to integrate into a determined mathematical model, so that the simulation experiment effect and the real physical experiment have great difference.
With the development and application of artificial intelligence algorithms such as machine learning and deep learning, when input and output and environmental data of the power system under a certain working condition are obtained, the input and output and environmental data can be used as a training data training model to predict the performance of the power system under the working condition. After the model is reasonably selected and the parameters are adjusted, the performance of the power system under the working condition can be predicted. However, a single machine learning and deep learning algorithm model is often poor in generalization capability, that is, when only training data under a conventional working condition is acquired, the model cannot be trained to predict the performance of the power system under some extreme working conditions. If training data under all working conditions, especially training data under some extreme working conditions, needs expensive experimental cost, even cannot be realized under laboratory conditions, and the performance of the power system under the extreme working conditions is often an important index for evaluating the reliability and stability of the power system.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a power equipment extreme condition performance prediction system based on Gaussian process regression.
The purpose of the invention can be realized by the following technical scheme:
a power plant extreme condition performance prediction system based on gaussian process regression, the system comprising:
the conventional working condition data acquisition unit: the system is used for dynamically acquiring system input, system output and environmental data of the power equipment in an experimental environment through the sensing equipment, namely conventional working condition data;
a terminal processing unit: the data transmission unit is used for receiving the data of the conventional working conditions and sending the data to the server unit through the data transmission unit;
a server unit: the method comprises the steps of receiving conventional working condition data, preprocessing the data to form a conventional working condition data set, constructing a Gaussian process regression model, training standard working conditions and extreme working conditions to obtain a trained Gaussian process regression model, feeding the trained Gaussian process regression model back to a terminal processing unit through a data transmission unit, manually inputting data through a human-computer interaction unit, predicting the performance of the extreme working conditions of the power equipment, and displaying a prediction result through the human-computer interaction unit.
The system input data of the power equipment are physical quantities which represent the running state, the system output data of the power equipment are physical quantities which represent the performance of the power equipment in the running process, the environmental data are temperature, atmospheric pressure and quantized vibration data in the running process of the power equipment, and the conventional working condition data are time series data and are sequences which change along with time in a certain time interval and are collected according to a set sampling frequency.
The method comprises the following steps of taking system input data and environment data as model inputs, taking system outputs as model outputs, and selecting different model inputs and outputs according to different power equipment and prediction projects, wherein the specific steps are as follows:
Figure BDA0003305076540000021
Figure BDA0003305076540000031
in the server unit, the data preprocessing unit is used for preprocessing the conventional working condition data, including singular value removal, smoothing and normalization processing, part of training data is randomly selected after the data form is arranged, a plurality of kernel functions are respectively adopted for fast fitting, the kernel function with the best fitting effect is selected as the kernel function of the Gaussian process regression model, and the kernel functions comprise SE, MA, RQ and SM.
In the server unit, the model pre-training unit is used for training the Gaussian process regression model under standard working conditions, and the method specifically comprises the following steps:
determining a negative logarithm marginal likelihood function according to training data and a kernel function to obtain an optimization problem which takes the negative logarithm marginal likelihood function as a target function and takes a hyperparameter as a decision variable, solving the optimization problem to obtain a proper hyperparameter, finishing the training of a Gaussian process regression model, and recording the Gaussian process regression model as GhAnd h is a hyper-parameter of the Gaussian process regression model.
In the server unit, the model extreme training unit is used for training the Gaussian process regression model under extreme working conditions, and the method specifically comprises the following steps:
1) initializing neural network model DωWherein, omega is a parameter of the neural network model, and the activation function of the output layer is a Sigmoid function;
2) sampling to obtain a DG function of the training data;
3) and executing a model extreme training algorithm to finish the training of extreme working conditions.
The step 2) specifically comprises the following steps:
21) respectively sampling batch times to obtain x-fxε to U (0,1), wherein fxFor the distribution obeyed by the input x,. epsilon.is a weight coefficient obtained from random sampling in the uniform distribution U (0, 1);
22) substituting training data input x into Gaussian process regression model GhIn (1), the upper and lower bounds p of the confidence interval are obtained±；
23) Respectively upper bound p of confidence interval+Substitution intoNeural network model DωTo obtain d1Lower bound p of confidence interval-Substitution neural network model DωTo obtain d2Wherein d is1、d2A process variable that is a DG function;
24) obtaining DG function of training data
Figure BDA0003305076540000032
The step 3) specifically comprises the following steps:
31) fixed Gaussian process regression model GhB, training a neural network DωParameter ω, in particular
Obtaining training data by using DG function, and inputting the training data into neural network model DωMiddle utilization neural network model DωUsing an Adam optimizer to find a neural network model D that minimizes lossDωParameter ω of (d);
32) fixed neural network model DωParameter omega, training a Gaussian process regression model GhThe hyper-parameter h specifically comprises:
using DG function to obtain training data, inputting into Gaussian process regression model GhIn (1), using a Gaussian process regression model GhUsing an Adam optimizer to find a Gaussian process regression model G that minimizes lossGhThe hyper-parameter h;
33) regression model G of Gaussian processhAnd returning the hyperparameter h to the terminal processing unit to obtain a function expression of the Gaussian process regression model.
In training Gaussian process regression model GhAnd when the input x acquired by the DG function is used, the data set of the conventional working condition is removed
Figure BDA0003305076540000041
The purpose of the part of the input data overlay is:
in the pre-training process, a Gaussian process regression model is trained to a conventional working condition data set
Figure BDA0003305076540000042
The data in the process has a good fitting effect, further training of extreme working conditions is carried out on the basis of ensuring the fitting effect of the part, and the value of the input value range parameter is adjusted according to the actual condition, so that the finally trained Gaussian process regression model obtains better temperature prediction capability under the extreme working conditions.
The human-computer interaction unit comprises a touch screen or consists of a display screen and physical keys, and an operator inputs model input data through the human-computer interaction unit, checks a prediction result output by the system to the model and changes parameters set manually.
Compared with the prior art, the performance prediction system and method of the power equipment extreme working condition based on the Gaussian process regression and the generated countermeasure network can be used for pre-training the Gaussian process regression model (the model can be used for predicting the performance of the power equipment under the conventional working condition) as training data after only obtaining the system input, the system output and the environmental data of the power equipment under the conventional working condition, and then further training the Gaussian process regression model through the model extreme training algorithm based on the Gaussian process regression and the generated countermeasure network, so that the performance of the power equipment under the extreme working condition, namely under the full working condition, can be predicted.
The invention has the following advantages:
1. the performance prediction of the power equipment under all working conditions including extreme working conditions can be realized, a manufacturer is helped to evaluate the reliability of the system, and operators in different regions and different use habits are helped to know the heating performance of the power equipment in an individualized way;
2. the experiment is carried out only under the conventional working condition, so that the experiment cost including manpower, material resources and time resources is greatly reduced, and the experiment difficulty is greatly reduced;
3. the Gaussian process regression model provides uncertainty analysis for the prediction result, such as variance and confidence interval of the prediction result, and has important significance in engineering application.
Drawings
FIG. 1 is a functional block diagram of a prediction system of the present invention.
FIG. 2 is a schematic diagram of a Gaussian process regression model (GPR) structure for a time series.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
Interpretation of terms:
a power plant: the device converts, conducts and adjusts various potential energy sources in the nature. In the production process of enterprises, the energy-saving machine system can convert the potential energy of nature into mechanical energy, then convert the mechanical energy into electric energy and convert the electric energy into the mechanical energy. The power equipment is divided into the following types according to different links in a power system:
1. a power generation device: such as steam boilers, steam engines, wok and camel machines, steam turbines, gasoline engines, diesel engines, generators and the like.
2. Power transmission and distribution equipment: such as transformers, switchboards, rectifiers, etc.
3. A power consuming device: such as electric motors, electric furnaces, electrolysis baths, air picks, electric welders, electrical appliances, etc.
4. D-axis and q-axis of the permanent magnet synchronous motor: a coordinate system is established on a motor rotor, the coordinate system and the rotor rotate synchronously, the direction of a rotor magnetic field is taken as a d axis, and the direction vertical to the rotor magnetic field is taken as a q axis.
5. Working conditions are as follows: the operating conditions of the power plant at a certain time may be embodied in the present invention as some combination of system input and environmental data collected in a conventional condition data collection unit.
6. Rated working condition: various technical indexes of the power equipment are in the running state under the rated state.
7. Designing a working condition: all technical indexes and states of the power equipment during operation meet the operation condition of the design requirement.
8. And (3) normal working conditions: including nominal and design conditions.
9. Extreme conditions: and the running state of each technical index and state of the power equipment beyond the range of the conventional working condition.
10. Performance: physical quantities that characterize the degree to which the power plant fulfills its function or the degree of negative influence that occurs when fulfilling its function, such as harmful emissions of diesel engines, oil consumption of gasoline engines, internal losses of electric motors, residual capacity of batteries, efficiency of transmission equipment, noise level of generators, vibration of steam turbines, etc.
As shown in FIG. 1, the invention provides a performance prediction system for extreme conditions of power equipment based on Gaussian process regression and generation of a countermeasure network, the system comprises a conventional condition data acquisition unit, a data transmission unit, a server unit, a human-computer interaction unit and a terminal processing unit, wherein the server unit comprises an input/output interface, a data preprocessing unit, a model pre-training unit and a model extreme training unit, and the system is specifically realized by the following steps:
step 1, a conventional working condition data acquisition unit dynamically acquires system input, system output and environment data of power equipment in an experimental environment by using sensing equipment, the system input, the system output and the environment data are collectively called as conventional working condition data, and the conventional working condition data are transmitted to a terminal processing unit;
step 2, the terminal processing unit sends the received normal working condition data to the server unit through the data transmission unit for calculation processing, and the method specifically comprises the following steps;
step 2.1, the data preprocessing unit preprocesses the acquired normal working condition data to obtain a normal working condition data set for training, and selects a kernel function of a Gaussian process regression model;
2.2, training the Gaussian process regression model by using a method of minimizing a negative log marginal likelihood function according to a standard Gaussian process regression model and an algorithm thereof by a model pre-training unit;
2.3, the model extreme training unit further trains the Gaussian process regression model in the step 2.2 by using a Gaussian process regression-based and confrontation network generation algorithm;
step 2.4, the server unit outputs the trained Gaussian process regression model and returns the model to the terminal processing unit through the data transmission unit;
step 3, the server unit feeds back the trained Gaussian process regression model to the terminal processing unit through the data transmission unit, waits for an operator to manually input data through the human-computer interaction unit, and executes step 4 if the human-computer interaction unit receives the manually input data; otherwise, continuing to wait in step 3;
and 4, substituting the manually input data into the Gaussian process regression model in the terminal processing unit to obtain a prediction result through calculation, and finally displaying the prediction result through the human-computer interaction unit.
In the above steps, the working condition refers to the running state of the power equipment at a certain moment, and in the invention, the working condition can be specifically a certain combination of the collected system input and the collected environmental data in the conventional working condition data collection unit; the rated working condition refers to the running condition of each technical index of the power equipment in a rated state; the design working condition refers to the operation condition that all technical indexes and states of the power equipment during operation meet the design requirements; the normal working conditions comprise rated working conditions and design working conditions; the extreme working condition refers to the running condition that each technical index and state of the power equipment exceed the range of the conventional working condition; the system input data refers to physical quantities representing the running state of the power equipment in the running process, such as the rotating speed, the torque, the current, the voltage and the like of the motor; the system output data refers to physical quantities representing the performance of the power equipment in the running process, such as temperature values of important parts (stators, windings, permanent magnets, rotors and the like) in the motor; the environmental data refers to the temperature, the atmospheric pressure, the quantized vibration condition and the like of the power equipment in the operation process. System input data and environment data are collectively called model input, and system output is called model output; different model inputs and outputs are selected according to different power equipment and prediction items, such as the example shown in table 1. Further, default values for the system input data and the environmental data should be set according to the actual configuration of the power plant and the environment of use.
TABLE 1 examples of input and output of power system performance prediction models
Figure BDA0003305076540000071
The normal operating condition data is time series data, i.e., system input, system output, and environmental data as a function of time. The normal working condition data for training is a sequence which is collected according to a set sampling frequency and changes along with time in a period of time interval, in other words, a time interval is discretized into a plurality of time points, and a group of system input, system output and environment data are collected at each time point.
The conventional working condition data acquisition unit uses a temperature sensor, a speed sensor, a torque sensor, a current and voltage sensor and the like as sensing equipment and is used for acquiring conventional working condition data.
In the server unit, the process of model training is as follows:
a data preprocessing unit:
firstly, preprocessing the data of the conventional working condition by using methods such as singular value removal, smoothing, normalization and the like, and arranging the data form: as shown in FIG. 2, at time t, the model input xtIs system input at time t, environmental data and system output data x at time t-1t-1Output of the model ytAnd outputting data for the system at the time t. In particular, at the initial time, x needs to be manually set according to the actual situationt-1. Thus, at each time point there is a set of training data for the model: (x)t,yt) (ii) a And obtaining how many groups of training data according to how many time points are taken.
Then calculate the model input data xtProbability density distribution fx。
And finally, randomly selecting a small part of training data from the preprocessed training data, keeping the original sequence, respectively and quickly fitting the training data by using a plurality of kernel functions (including but not limited to the kernel functions shown in the table 2), and selecting the kernel function with the best fitting effect as the kernel function used in the Gaussian process regression model training unit. After the kernel function is determined, the number of the hyper-parameters in the Gaussian process regression model can be determined. The term "best fit" means using the kernel functionObtained by line fitting
Figure BDA0003305076540000081
At a minimum, wherein: y isiIn order to be the true value of the value,
Figure BDA0003305076540000082
for the prediction values, n is the total number of training data participating in the fast fit.
TABLE 2 Kernel functions commonly used in Gaussian Process regression models
Figure BDA0003305076540000083
A model pre-training unit:
the gaussian process regression model is trained using a standard gaussian process regression model and its algorithm. Training a Gaussian process regression model, namely determining the value of the hyperparameter in the model: firstly, determining a negative logarithm marginal likelihood function according to training data and a kernel function; the negative log marginal likelihood function is a function of a hyper-parameter, and the number of the hyper-parameter is determined according to the kernel function selected in the step 2.1, so that an optimization problem which takes the negative log marginal likelihood function as a target function and the hyper-parameter as a decision variable is obtained; solving the optimization problem can obtain proper hyper-parameters to complete the training of the regression model of the Gaussian process. The Gaussian process regression model is denoted as GhAnd h is a hyper-parameter of the Gaussian process regression model.
The mean function and covariance function can be obtained by training the gaussian process regression model, as shown in table 3. The mean function fits the functional relationship between the model input and the output, that is, the known model input is substituted into the mean function, so that the predicted output of the Gaussian process regression model about the input can be obtained. Further, in table 3, the predicted variance and confidence interval can be obtained from the mean function and covariance function.
TABLE 3 calculation formula of main function in regression model of Gaussian process
Figure BDA0003305076540000091
Model extreme training unit:
first, a neural network model D is initializedωWhere ω is a parameter of the neural network model. DωThe structure of the method is designed according to the actual situation, and the activation function of the output layer is guaranteed to be a Sigmoid function;
algorithm 2 is then executed as shown in table 4. In table 4, algorithm 1 is a secondary algorithm to algorithm 2. In Algorithm 2, the Gaussian process regression model G is first fixedhB, training a neural network DωParameter omega (phi-phi): obtaining training data using DG function, input DωMiddle utilization of DωThe output of (2) constitutes a loss function lossD; using Adam optimizer, find neural network D that minimizes lossDωParameter ω of (d). Then fix the neural network DωParameter omega, training a Gaussian process regression model GhOver parameter h of
Figure BDA0003305076540000092
Using DG function to obtain training data, input GhIn using GhThe output of (1) constitutes a loss function lossG; using Adam optimizer, find a Gaussian process regression model G that minimizes lossGhThe hyperparameter h. Wherein
Figure BDA0003305076540000101
A framework for generating a countermeasure network is used. With a Gaussian process regression model GhThe hyperparameter h of (1) can be used for obtaining each function expression of the Gaussian process regression model according to the table 3.
TABLE 4 Algorithm Table
Figure BDA0003305076540000102
Figure BDA0003305076540000111
In particular, it is recommended to train GhWhen the X acquired by the DG function is used, the data set of the conventional working condition is properly rejected
Figure BDA0003305076540000112
The input data overlay. For example when
Figure BDA0003305076540000113
Is one-dimensional and the range of the input is [ -2,2]Then, x used for training should satisfy: x to fxAnd | x | is less than or equal to 2 · c, wherein 0 is less than or equal to c is less than or equal to 1, the purpose of doing so is: in pre-training, the Gaussian process regression model pairs have been trained
Figure BDA0003305076540000114
The data in (1) has good fitting effect; further training should be performed on the basis of a moderate guarantee of the fitting effect of this part. And adjusting the value of c according to the actual condition, so that the finally trained Gaussian process regression model can obtain better temperature prediction capability under the extreme working condition.
Training Gaussian process regression model GhIs the process of determining its hyper-parameter h, so for a trained GhIn step 2.4, only the hyper-parameter h needs to be returned to the terminal processing unit.
The man-machine interaction unit consists of a touch screen or a display screen and physical keys. An operator can input model input data through the human-computer interaction unit, check the prediction result of the system on the output of the model (the temperature of a key part in the motor) and change parameters needing to be set manually in the algorithm.
Examples
The fuel consumption prediction system for the vehicle/ship fuel engine under the extreme working condition based on the Gaussian process regression and the generation countermeasure network disclosed by the embodiment is used for predicting the fuel consumption rate (L/100KM) of the vehicle/ship fuel engine under the full working condition, particularly under the extreme working condition. As described in detail below.
The conventional working condition data acquisition unit: by way of example and not limitation, the data acquisition unit may be a temperature sensor, a gravity sensor, a speed sensor, a torque sensor, a pressure sensor, a real-time detection detector for vehicle fuel consumption, and the like directly mounted on the power system device, as a sensing device, and is configured to acquire input (required rotation speed, required torque, gear, driving mode, vehicle load, tire pressure, total 6 items), output (real-time fuel consumption (L/100KM), total 1 item) and environmental data (temperature, atmospheric pressure of the vehicle power system, and usage of quantized vehicle-mounted non-power system electrical equipment, road flatness, total 4 items) of the power system; at the initial moment, the real-time oil consumption is 0. And sending the conventional working condition data to the terminal processing unit through a WIFI network, an AP hotspot or other transmission modes.
A terminal processing unit: by way of example and not limitation, the system mainly comprises a storage medium and a processor, wherein the processor is configured to execute a program on the storage medium, and the program on the storage medium performs necessary transcoding on the acquired normal operating condition data, such as converting an electrical signal into a digital signal.
A data transmission unit: by way of example and not limitation, the TCP/IP protocol may be used to perform data transfer between different units in a temperature prediction system based on gaussian process regression and the extreme operating conditions of the motor that generate the countermeasure network.
A server unit: the structure of the gaussian process regression model for time series is shown in figure two.
Firstly, the normal working condition data is preprocessed by normalization, singular value removal, smoothing and the like to obtain a training data set with the capacity of n
Figure BDA0003305076540000121
Wherein the content of the first and second substances,
Figure BDA0003305076540000122
the vector is input data of the motor at the ith moment and output data of the motor at the (i-1) th moment, namely input of the model; y isiThe output data of the battery at the ith moment is also the output of the model; specifically, at the time when i is 0, the "output data of the battery at the time i-1" is regarded as the ambient temperatureThe value is obtained. From the training data set
Figure BDA0003305076540000123
In the random selection
Figure BDA0003305076540000124
(if not, rounding down) a group of training data, fitting the training data by using a Gaussian process regression model (attached table I) and different kernel functions (attached table II) respectively to find the best fitting effect, namely
Figure BDA0003305076540000125
The smallest kernel function, which is taken as the selected kernel function. Finally, the training data set is obtained
Figure BDA00033050765400001222
Using a non-parametric distribution fitting method to calculate the probability density distribution fx。
Selecting kernel function, namely determining the number of hyperparameters, writing out negative logarithmic marginal likelihood function by combining with Bayesian theorem, namely the property of Gaussian random process, selecting SE (squared explicit) kernel function as an example and not limitation, wherein the negative logarithmic marginal likelihood function is
Figure BDA0003305076540000126
Wherein the content of the first and second substances,
Figure BDA0003305076540000127
is composed of yiE R, i 1.., n, K is a matrix of n × n:
Figure BDA0003305076540000128
i is an identity matrix of n × n, σnIs a hyper-parameter. Meanwhile, the K matrix also comprises two other hyper-parameters (attached table II). Order to
Figure BDA0003305076540000129
Representing a vector consisting of three hyper-parameters, solving the optimization problem using particle swarm optimization
Figure BDA00033050765400001210
The most suitable hyper-parameter can be obtained
Figure BDA00033050765400001211
The feasible domain of the hyper-parameter can be limited according to the actual situation, namely:
Figure BDA00033050765400001212
then, using the algorithm 2 in the attached table IV, a Gaussian process regression model G is obtained through calculationhThe hyperparameter h is calculated according to the attached table three to obtain the mean function of the Gaussian process regression model
Figure BDA00033050765400001213
I.e. input into the model as
Figure BDA00033050765400001214
Time, model output
Figure BDA00033050765400001215
Is described in (1).
In actual prediction, a set of time series model inputs consisting of input and output of the fuel engine and environmental data are given:
Figure BDA00033050765400001216
(wherein is complete)
Figure BDA00033050765400001217
Need to firstly
Figure BDA00033050765400001223
Substitution into
Figure BDA00033050765400001218
The model output at the time when t is 0 can be obtained,
Figure BDA00033050765400001219
like), in turn, substitute
Figure BDA00033050765400001220
Calculating to obtain model output, namely a prediction result: time series of changes in oil consumption (L/100 KM):
Figure BDA00033050765400001221
further, the average oil consumption in the 0-t time period is:
Figure BDA0003305076540000131
the sampling interval of the time series should be as small as possible.
A human-computer interaction unit: by way of example and not limitation, may consist of a touch screen or a display screen and buttons. The user completes the input of the required checking working condition and various system parameters by using the man-machine interaction unit, and obtains the prediction result information of the relevant operation.
The fuel consumption prediction system for the vehicle/ship fuel engine under the extreme working condition based on Gaussian process regression and generation countermeasure network comprises the following use flow:
1. installing sensing equipment at a corresponding position of the fuel engine;
2. starting the fuel engine to enable the fuel engine to operate under a certain or a plurality of standard working conditions or other working conditions which are easy to reach in laboratories and real working environments;
3. the fuel engine is closed, after the calculation of the server unit is completed, the working condition data to be checked, namely the input data (including system input and environment data, which are a time sequence) of the fuel engine model is input through the man-machine interaction unit;
4. and checking the prediction result through a human-computer interaction unit.

Claims (10)

1. A power equipment extreme condition performance prediction system based on Gaussian process regression is characterized by comprising the following components:
the conventional working condition data acquisition unit: the system is used for dynamically acquiring system input, system output and environmental data of the power equipment in an experimental environment through the sensing equipment, namely conventional working condition data;
a terminal processing unit: the data transmission unit is used for receiving the data of the conventional working conditions and sending the data to the server unit through the data transmission unit;
a server unit: the method comprises the steps of receiving conventional working condition data, preprocessing the data to form a conventional working condition data set, constructing a Gaussian process regression model, training standard working conditions and extreme working conditions to obtain a trained Gaussian process regression model, feeding the trained Gaussian process regression model back to a terminal processing unit through a data transmission unit, manually inputting data through a human-computer interaction unit, predicting the performance of the extreme working conditions of the power equipment, and displaying a prediction result through the human-computer interaction unit.
2. The system for predicting the performance of the power equipment under the extreme operating conditions based on the gaussian process regression is characterized in that system input data of the power equipment are physical quantities which represent the operating state, system output data of the power equipment are physical quantities which represent the performance of the power equipment in the operating process, environmental data are temperature, atmospheric pressure and quantized vibration data in the operating process of the power equipment, and the data of the conventional operating conditions are time series data and are sequences which change along with time within a time interval and are collected according to a set sampling frequency.
3. The system for predicting the extreme condition performance of the power equipment based on the Gaussian process regression as claimed in claim 2, is characterized in that system input data and environment data are used as model inputs, system outputs are used as model outputs, and different model inputs and outputs are selected according to different power equipment and prediction items, specifically:
Figure FDA0003305076530000011
Figure FDA0003305076530000021
4. the system for predicting the extreme condition performance of the power equipment based on the gaussian process regression as claimed in claim 2, wherein the server unit preprocesses the normal condition data through the data preprocessing unit, wherein the preprocessing includes singular value removal, smoothing and normalization, and randomly selects part of training data after data form arrangement, and respectively adopts a plurality of kernel functions to perform fast fitting, and selects the kernel function with the best fitting effect as the kernel function of the gaussian process regression model, and the kernel functions include SE, MA, RQ and SM.
5. The system for predicting the extreme condition performance of the power equipment based on the gaussian process regression as claimed in claim 4, wherein the server unit performs standard condition training on the gaussian process regression model through the model pre-training unit, and specifically comprises:
determining a negative logarithm marginal likelihood function according to training data and a kernel function to obtain an optimization problem which takes the negative logarithm marginal likelihood function as a target function and takes a hyperparameter as a decision variable, solving the optimization problem to obtain a proper hyperparameter, finishing the training of a Gaussian process regression model, and recording the Gaussian process regression model as GhAnd h is a hyper-parameter of the Gaussian process regression model.
6. The system for predicting the extreme condition performance of the power equipment based on the gaussian process regression as claimed in claim 4, wherein the server unit trains the gaussian process regression model for the extreme condition through the model extreme training unit, specifically:
1) initializing neural network model DωWherein, omega is a parameter of the neural network model, and the activation function of the output layer is a Sigmoid function;
2) sampling to obtain a DG function of the training data;
3) and executing a model extreme training algorithm to finish the training of extreme working conditions.
7. The system for predicting the extreme condition performance of the power equipment based on the Gaussian process regression as claimed in claim 6, wherein the step 2) comprises the following steps:
21) respectively sampling batch times to obtain x-fxε to U (0,1), wherein fxFor the distribution obeyed by the input x,. epsilon.is a weight coefficient obtained from random sampling in the uniform distribution U (0, 1);
22) substituting training data input x into Gaussian process regression model GhIn (1), the upper and lower bounds p of the confidence interval are obtained±；
23) Respectively upper bound p of confidence interval+Substitution neural network model DωTo obtain d1Lower bound p of confidence interval-Substitution neural network model DωTo obtain d2Wherein d is1、d2A process variable that is a DG function;
24) obtaining DG function of training data
Figure FDA0003305076530000031
8. The system for predicting the extreme condition performance of the power equipment based on the Gaussian process regression as claimed in claim 6, wherein the step 3) comprises the following steps:
31) fixed Gaussian process regression model GhB, training a neural network DωParameter ω, in particular
Obtaining training data by using DG function, and inputting the training data into neural network model DωMiddle utilization neural network model DωUsing an Adam optimizer to find a neural network model D that minimizes lossDωParameter ω of (d);
32) fixed neural network model DωParameter omega, training a Gaussian process regression model GhThe over-parameter h of (a) is,the method specifically comprises the following steps:
using DG function to obtain training data, inputting into Gaussian process regression model GhIn (1), using a Gaussian process regression model GhUsing an Adam optimizer to find a Gaussian process regression model G that minimizes lossGhThe hyper-parameter h;
33) regression model G of Gaussian processhAnd returning the hyperparameter h to the terminal processing unit to obtain a function expression of the Gaussian process regression model.
9. The system of claim 8, wherein the regression model G is trained during the Gaussian processhAnd when the input x acquired by the DG function is used, the data set of the conventional working condition is removed
Figure FDA0003305076530000032
The purpose of the part of the input data overlay is:
in the pre-training process, a Gaussian process regression model is trained to a conventional working condition data set
Figure FDA0003305076530000033
The data in the process has a good fitting effect, further training of extreme working conditions is carried out on the basis of ensuring the fitting effect of the part, and the value of the input value range parameter is adjusted according to the actual condition, so that the finally trained Gaussian process regression model obtains better temperature prediction capability under the extreme working conditions.
10. The system for predicting the extreme condition performance of the power equipment based on the Gaussian process regression as claimed in claim 1, wherein the human-computer interaction unit comprises a touch screen or consists of a display screen and physical keys, an operator inputs model input data through the human-computer interaction unit, checks a prediction result output by the system to the model, and changes parameters set manually.
CN202111201676.3A 2021-10-15 2021-10-15 Power equipment extreme condition performance prediction system based on Gaussian process regression Pending CN114004391A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111201676.3A CN114004391A (en) 2021-10-15 2021-10-15 Power equipment extreme condition performance prediction system based on Gaussian process regression

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111201676.3A CN114004391A (en) 2021-10-15 2021-10-15 Power equipment extreme condition performance prediction system based on Gaussian process regression

Publications (1)

Publication Number Publication Date
CN114004391A true CN114004391A (en) 2022-02-01

Family

ID=79922927

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111201676.3A Pending CN114004391A (en) 2021-10-15 2021-10-15 Power equipment extreme condition performance prediction system based on Gaussian process regression

Country Status (1)

Country Link
CN (1) CN114004391A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114662252A (en) * 2022-02-25 2022-06-24 佳木斯大学 Method for improving performance index of complex networked random system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114662252A (en) * 2022-02-25 2022-06-24 佳木斯大学 Method for improving performance index of complex networked random system

Similar Documents

Publication Publication Date Title
CN111460379B (en) Multi-working-condition power system performance prediction method and system based on Gaussian process regression
Hang et al. Fault diagnosis of wind turbine based on multi‐sensors information fusion technology
CN111413619B (en) Multi-working-condition battery residual capacity prediction method and system based on Gaussian process regression
Fan et al. A novel machine learning method based approach for Li-ion battery prognostic and health management
Rao et al. A deep bi-directional long short-term memory model for automatic rotating speed extraction from raw vibration signals
CN113469470B (en) Energy consumption data and carbon emission correlation analysis method based on electric brain center
CN111460382A (en) Fuel vehicle harmful gas emission prediction method and system based on Gaussian process regression
Hou et al. Diagnosisformer: An efficient rolling bearing fault diagnosis method based on improved Transformer
CN114004391A (en) Power equipment extreme condition performance prediction system based on Gaussian process regression
CN111460380A (en) Multi-working-condition driving range prediction method and system based on Gaussian process regression
Sheikh et al. A data-driven comparative analysis of lithium-ion battery state of health and capacity estimation
CN114019371B (en) Temperature prediction system for extreme working condition of motor based on Gaussian process regression
Yang et al. Physics-informed multi-state temporal frequency network for RUL prediction of rolling bearings
Jiang et al. Two‐level area‐load modelling for OPF of power system using reinforcement learning
Wang et al. Probability estimation based on grey system theory for simulation evaluation
Liu et al. Fuzzy equivalence relation clustering-based algorithm for coherency identification among generators
Li et al. Intelligent prediction on air intake flow of spark ignition engine by a chaos radial basis function neural network
Xu et al. Intelligent fault identification in sample imbalance scenarios using robust low-rank matrix classifier with fuzzy weighting factor
Channegowda et al. An attention‐based synthetic battery data augmentation technique to overcome limited dataset challenges
CN113835964A (en) Cloud data center server energy consumption prediction method based on small sample learning
Meng et al. Study on immune cloud model adapted to complex mechanical system diagnosis
Zhao et al. A deep neural networks based on multi-task learning and its application
Shen et al. Design of MultiChannel Parallel Dynamic Testing Virtual Instrument Based on Dynamic Measurement
Ren et al. A two-stage fuzzy nonlinear combination method for utmost-short-term wind speed prediction based on TS fuzzy model
Zhang et al. Fault judgment of transmission cable based on multi-channel data fusion and transfer learning

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