Generation quantity acquisition method and device, storage medium and electronic device

Abstract

The invention discloses a power generation quantity acquisition method and device, a storage medium and an electronic device, and the method comprises the steps: analyzing the product content of an oxidation product of fan lubricating oil configured for a power generation fan through a neural network model, and obtaining an oxidation resistance parameter value corresponding to the fan lubricating oil, the neural network model is obtained by training by taking the historical product content of historical oxidation products generated when the power generation fan works by using historical fan lubricating oil as an input sample and taking a historical oxidation resistance parameter value corresponding to the historical fan lubricating oil as an output sample; target fan lubricating oil with the maximum oxidation resistance parameter value is determined from the multiple kinds of fan lubricating oil according to the multiple kinds of fan lubricating oil; and target fan lubricating oil is configured on the power generation fan, and the power generation quantity generated when the power generation fan works through the target fan lubricating oil is obtained. The problem that the power generation quantity is insufficient due to the fact that lubricating oil not high enough in oxidation resistance is used for power generation work of a power generation fan is solved.

Description

Technical Field

[0001] The present application relates to the technical field of power generation of wind turbines, and more specifically, to a method and device for acquiring power generation amount, a storage medium and an electronic device. Background Art

[0002] The wind turbine is a highly automated energy conversion mechanical system that places extremely high demands on the operating accuracy and stability of the mechanical equipment. When the wind turbine is working, the friction parts of the transmission system of the wind turbine are in frequent start-stop and high-load continuous operation conditions, which will cause mechanical consumption of the wind turbine. Although lubricating oil can be added to the mechanical contacts of various transmission gears and bearings of the wind turbine to form an oil film on the contact surface to reduce friction and wear and extend the service life of the machinery, the oxidation of the lubricating oil can easily cause the viscosity of the oil to increase, and non-oil-soluble polymers to form deposits such as paint films. Therefore, it is often necessary to provide the wind turbine with lubricating oil with high antioxidant properties.

[0003] The common test methods for the oxidation resistance of lubricating oils are not accurate enough in analyzing the oxidation resistance of lubricating oils due to their complex composition and structure. The use of lubricating oils with insufficient oxidation resistance to generate electricity in wind turbines will affect the amount of electricity generated.

[0004] In the related art, there is no effective solution to the problem of insufficient power generation due to the use of lubricating oil with insufficient antioxidant properties for power generation in wind turbines.

[0005] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0006] The embodiments of the present application provide a method and device for obtaining power generation, a storage medium and an electronic device, so as to at least solve the problem of insufficient power generation caused by the use of lubricating oil with insufficient antioxidant properties for power generation in the power generation fan.

[0007] According to one aspect of an embodiment of the present application, a method for obtaining power generation is provided, comprising: using a neural network model to analyze the product content of oxidation products of fan lubricating oil configured for a power-generating wind fan, and obtaining an antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using historical product contents of historical oxidation products generated when the power-generating wind fan uses historical fan lubricating oil as input samples, and using historical antioxidant parameter values ​​corresponding to the historical fan lubricating oil as output samples; for multiple fan lubricating oils, determining a target fan lubricating oil with a maximum antioxidant parameter value from the multiple fan lubricating oils; configuring the target fan lubricating oil on the power-generating wind fan, and obtaining the power generation generated when the power-generating wind fan uses the target fan lubricating oil.

[0008] In an exemplary embodiment, the neural network model includes at least a radial basis function neural network model, and the antioxidant parameter value corresponding to the fan lubricant is obtained. The neural network model is used to analyze the product content of the oxidation products of the fan lubricant configured for the power generation wind turbine, including: setting an input layer, a hidden layer and an output layer in the radial basis function neural network model in sequence; inputting the product content of the oxidation product into the input layer, and inputting the first feature output by the input layer into the hidden layer, so as to use the hidden layer to perform a nonlinear transformation on the first feature using a radial basis function to obtain a second feature; using the output layer to perform a linear weighted operation on the second feature, and outputting the antioxidant parameter value corresponding to the fan lubricant.

[0009] In an exemplary embodiment, the first feature output by the input layer is input to the hidden layer, so as to use the hidden layer to perform a nonlinear transformation on the first feature using a radial basis function to obtain a second feature, including: determining a radial basis center of the hidden layer, calculating thresholds corresponding to M neurons in the hidden layer, and calculating the second feature through the radial basis center;

[0010] The radial basis center of the hidden layer is expressed as: C=X', X is the input matrix corresponding to the first feature,

[0011] In an exemplary embodiment, the output layer is used to perform a linear weighted operation on the second feature to output the antioxidant parameter value corresponding to the fan lubricant, including: obtaining the connection weight between the hidden layer and the output layer, and setting an output layer threshold for the output layer; using the connection weight, the output layer threshold and the first feature to calculate the antioxidant parameter value.

[0012] In an exemplary embodiment, the matrix Y of the antioxidant parameter values ​​is represented by the following formula:

[0013] [WB2]·[AI]=Y;

[0014] Wherein, W is the connection weight, w ij represents the connection weight between the jth hidden neuron and the ith output neuron, B2 is the output layer threshold, B2 = [b 21 ,b 22 ,......b 2N ]', N is the number of output layers, A is the first feature, I = [1,1…1] 1×M .

[0015] In an exemplary embodiment, before using a neural network model to analyze the product content of oxidation products of fan lubricating oil configured for a power-generating wind turbine, the method further includes: using an infrared spectrometer to perform an oxidation test on the fan lubricating oil configured for the power-generating wind turbine to obtain oxidation band data of the fan lubricating oil, wherein the oxidation band data at least includes characteristic band data of different oxidation products generated by the fan lubricating oil when an oxidation reaction occurs; performing data cleaning on the oxidation band data to obtain clean data after cleaning; and determining the product content of the different oxidation products from the clean data.

[0016] According to another aspect of an embodiment of the present application, a device for acquiring power generation is also provided, including: an acquisition module, used to analyze the product content of oxidation products of fan lubricating oil configured for a power-generating wind fan by using a neural network model, and obtain an antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using the historical product content of historical oxidation products generated when the power-generating wind fan uses the historical fan lubricating oil as an input sample, and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample; a determination module, used to determine a target fan lubricating oil with a maximum antioxidant parameter value from among a plurality of fan lubricating oils; and an acquisition module, used to configure the target fan lubricating oil on the power-generating wind fan, and obtain the power generation generated when the power-generating wind fan uses the target fan lubricating oil.

[0017] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned method for obtaining the generated electricity when running.

[0018] According to another aspect of the embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for obtaining the generated electricity through the computer program.

[0019] According to another aspect of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0020] Through the present application, the product content of the oxidation products of the fan lubricant configured for the power generation wind turbine is analyzed by using a neural network model to obtain the antioxidant parameter value corresponding to the fan lubricant, wherein the neural network model is trained using the historical product content of the historical oxidation products generated when the power generation wind turbine uses the historical fan lubricant as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricant as an output sample; for multiple fan lubricants, a target fan lubricant with the maximum antioxidant parameter value is determined from the multiple fan lubricants; the target fan lubricant is configured on the power generation wind turbine, and the power generation generated when the power generation wind turbine uses the target fan lubricant is obtained. Under the premise of knowing the type of antioxidant, the content of the oxidation products can be analyzed based on the neural network model to evaluate the antioxidant performance of the lubricant, and a fan lubricant with high antioxidant performance can be selected for the power generation wind turbine, thereby solving the problem of insufficient power generation due to the use of lubricants with insufficient antioxidant properties for power generation in the power generation wind turbine, improving the power generation efficiency, and thereby increasing the power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0023] Figure 1 It is a hardware structure block diagram of a mobile terminal of a method for obtaining power generation amount according to an embodiment of the present application;

[0024] Figure 2 is a flow chart of a method for obtaining power generation according to an embodiment of the present application;

[0025] Figure 3 is a flowchart of a training process of a neural network model according to an embodiment of the present application;

[0026] Figure 4 It is a structural block diagram of a device for acquiring power generation according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal of a method for obtaining power generation in an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0030] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for obtaining the power generation in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In this embodiment, a method for obtaining generated electricity is provided. Figure 2 is a flow chart of a method for obtaining power generation according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0033] Step S202: using a neural network model to analyze the product content of oxidation products of a fan lubricating oil (hereinafter referred to as lubricating oil) configured for a power-generating fan, and obtaining an antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using a historical product content of historical oxidation products generated when the power-generating fan uses the historical fan lubricating oil as an input sample, and a historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample;

[0034] Optionally, the neural network model is, for example, an RBF (Radial Basis Function Neural Network) model. The RBF neural network model is a feedforward neural network that uses a radial basis function as an activation function of a hidden layer. It has a one-way forward propagation process, and the training and learning speed of the training samples is fast. After the training and learning is completed, the weights and thresholds between the layers of the RBF neural network model are determined accordingly, and there is no need for repeated iterations to adjust the weights and thresholds.

[0035] Among them, the antioxidant parameter value is used as an evaluation standard to measure the antioxidant performance. The larger the antioxidant parameter, the less the content of oxidation products.

[0036] Step S204: for a plurality of fan lubricating oils, determining a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils;

[0037] Step S206: configuring the target wind turbine lubricant on the power-generating wind turbine, and obtaining the amount of power generated when the power-generating wind turbine operates using the target wind turbine lubricant.

[0038] Through the above steps, the product content of the oxidation products of the fan lubricant configured for the power generation wind turbine is analyzed by using the neural network model to obtain the antioxidant parameter value corresponding to the fan lubricant, wherein the neural network model is trained using the historical product content of the historical oxidation products generated when the power generation wind turbine uses the historical fan lubricant as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricant as an output sample; for multiple fan lubricants, a target fan lubricant with the maximum antioxidant parameter value is determined from the multiple fan lubricants; the target fan lubricant is configured on the power generation wind turbine, and the power generation generated when the power generation wind turbine uses the target fan lubricant is obtained. Under the premise of knowing the type of antioxidant, the content of the oxidation products can be analyzed based on the neural network model to evaluate the antioxidant performance of the lubricant, and a fan lubricant with high antioxidant performance can be selected for the power generation wind turbine, thereby solving the problem of insufficient power generation due to the use of lubricants with insufficient antioxidant properties for power generation in the power generation wind turbine, improving the power generation efficiency, and thereby increasing the power generation.

[0039] In an exemplary embodiment, the neural network model includes at least a radial basis function neural network model. The process of analyzing the product content of the oxidation products of the fan lubricant configured for the power generation wind turbine using the neural network model for the antioxidant parameter value corresponding to the fan lubricant obtained in the above step S202 may include: sequentially setting an input layer, a hidden layer, and an output layer in the radial basis function neural network model; inputting the product content of the oxidation products into the input layer, and inputting the first feature output by the input layer into the hidden layer, so as to use the hidden layer to perform a nonlinear transformation on the first feature using a radial basis function to obtain a second feature; and using the output layer to perform a linear weighted operation on the second feature to output the antioxidant parameter value corresponding to the fan lubricant.

[0040] The RBF neural network model generally includes an input layer, a hidden layer, and an output layer. In the RBF neural network model, the activation function of the hidden layer neurons is a radial basis function, usually a Gaussian function. The characteristic of the radial basis function is that it has a local response characteristic, that is, when the input data is close to the central value of a neuron, the output of the neuron is large; conversely, when the input data is greatly different from the central value, the output of the neuron is small.

[0041] The connection between the input layer and the hidden layer is unweighted, and there is no weight matrix. The input data can be directly passed to the hidden layer neurons without weighting. Alternatively, the connection between the input layer and the hidden layer can be regarded as a connection with a weight of 1.

[0042] The hidden layer uses a nonlinear optimization strategy to adjust the parameters of the activation function. Each hidden layer neuron has a center point and a width parameter. The center point determines the response area of ​​the neuron, and the width parameter controls the response range of the neuron. The transformation from the input layer space to the hidden layer space is a nonlinear transformation for feature extraction, while the transformation from the hidden layer space to the output layer space is a linear transformation.

[0043] A linear activation function is usually used in the output layer to perform a linear transformation on the input matrix of the output layer (for example, outputting the input variables through a purelin function linear weighting) to generate the final output result.

[0044] This application can better reflect the complex nonlinear relationship between the input and output of the research target by using the RBF neural network model, establish a functional relationship between the content of lubricating oil oxidation products and the anti-oxidation property of lubricating oil, and quantitatively judge the changes in the anti-oxidation performance of oil products.

[0045] For the RBF neural network model, the training process includes two stages: determining the center point and width parameters (usually using unsupervised learning methods such as clustering algorithms), and adjusting the weights and biases (using supervised learning methods such as the least squares method) to make the network's predicted output as close to the true value as possible. The trained RBF neural network model has the advantages of global approximation ability, strong generalization ability, and fast training speed. It is suitable for function approximation, pattern recognition, classification, and time series prediction. It can be applied in signal processing, image recognition, financial prediction and other fields.

[0046] Furthermore, through pure linear weighted sum, the RBF network can have good generalization ability, that is, it can achieve good prediction results on data outside the training set.

[0047] In an exemplary embodiment, it is further proposed that the first feature output by the input layer is input into the hidden layer, so as to use the hidden layer to perform nonlinear transformation on the first feature using a radial basis function to obtain a second feature, which specifically includes: determining the radial basis center of the hidden layer, calculating the thresholds corresponding to M neurons in the hidden layer, and calculating the second feature through the radial basis center;

[0048] The radial basis center of the hidden layer is expressed as: C=X', X is the input matrix corresponding to the first feature, p, M are positive integers.

[0049] Optionally, C represents the center point of the radial basis function, which is essentially a p×M matrix in which each row represents a center point. In the radial basis function neural network, each center point corresponds to a radial basis function. The distance between the input feature vector X and these center points C will be used to calculate the output of each radial basis function. And X' can be understood as X after some transformation, such as transposition, standardization or other preprocessing methods. If X' directly represents the transpose of X, then it will be an n×m matrix in which each row represents a feature and each column represents a sample.

[0050] In an exemplary embodiment, the process of using the output layer to perform a linear weighted operation on the second feature and outputting the antioxidant parameter value corresponding to the fan lubricating oil is further described by the following steps: obtaining the connection weight between the hidden layer and the output layer, and setting an output layer threshold for the output layer; using the connection weight, the output layer threshold and the first feature to calculate the antioxidant parameter value.

[0051] Optionally, the linear weighting operation on the second feature using the output layer can be implemented using, for example, a "purelin" function. When using a radial basis function (RBF) neural network, the "purelin" function generally refers to a linear activation function used in the output layer.

[0052] Here "purelin" is a term that means a purely linear function, that is, a simple weighted sum without nonlinear transformation. Each neuron in the hidden layer corresponds to a radial basis function, usually a Gaussian function. These functions perform nonlinear transformations on the input data and generate a set of new hidden layer outputs, which can be regarded as some kind of representation of the input data in the feature space. In the output layer, the activation function of the neurons is purely linear, that is, there is no nonlinear transformation. This means that the neurons in the output layer directly perform a weighted sum of the outputs of the hidden layer without any nonlinear processing. Each neuron in the output layer receives the outputs from all hidden layer neurons and then performs a linear weighted sum on these outputs. The final output of the output layer neuron is the result of the weighted sum plus the bias term. This result can be a continuous value or a discrete value, depending on the nature of the specific problem.

[0053] Among them, weights are parameters learned by the RBF network during the training process, which represent the connection strength from the hidden layer to the output layer. In each neuron in the output layer, in addition to the weighted sum, there can also be a bias term. The bias term is a constant that can adjust the output value of the output layer neuron.

[0054] In short, in the RBF neural network, the process of using purelin function to perform linear weighting and output of input variables is the key link for the network to achieve nonlinear feature extraction and linear prediction. By reasonably designing the network structure and training process, the RBF network can solve various complex nonlinear problems.

[0055] In an exemplary embodiment, the matrix Y of the antioxidant parameter values ​​can be represented by the following formula:

[0056] [WB2]·[AI]=Y;

[0057] Wherein, W is the connection weight, w ij represents the connection weight between the jth hidden neuron and the ith output neuron, B2 is the output layer threshold, B2 = [b 21 ,b 22 ,......b 2N ]', N is the number of output layers, A is the first feature, I = [1,1…1] 1×M .

[0058] In an exemplary embodiment, before using a neural network model to analyze the product content of oxidation products of fan lubricating oil configured for a power-generating wind turbine, an infrared spectrometer can be further used to perform an oxidation test on the fan lubricating oil configured for the power-generating wind turbine to obtain oxidation band data of the fan lubricating oil, wherein the oxidation band data at least includes characteristic band data of different oxidation products generated by the fan lubricating oil when an oxidation reaction occurs; the oxidation band data is cleaned to obtain clean data after cleaning; and the product content of the different oxidation products is determined from the clean data.

[0059] Through this embodiment, the content change of the oxidation product generated during the oxidation process of the lubricating oil can be analyzed, and then the change of the anti-oxidation performance of the oil can be determined. That is, based on the known types of antioxidants, the anti-oxidation performance of the lubricating oil can be determined by measuring the characteristic absorption peaks of the antioxidants.

[0060] Obviously, the embodiments described above are only some embodiments of the present application, not all embodiments. In order to better understand the above method, the training process of the above neural network model is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present application, specifically:

[0061] In an optional embodiment, the present application Figure 3 The following steps are explained:

[0062] Step S1, performing infrared testing on lubricating oil samples, specifically collecting lubricating oil samples to be tested, using an infrared spectrometer to perform testing and analysis, and obtaining full-band information in the infrared spectrum.

[0063] Infrared testing can identify the molecular structure of unknown substances or determine their chemical groups, and conduct quantitative analysis and purity identification. Moreover, the detection process is not limited by the phase of the sample, and solid, liquid or gas samples can be directly measured. Infrared spectroscopy requires less sample volume, fast analysis speed, and easy operation. Therefore, in the study of the antioxidant properties of lubricating oils, the antioxidant properties of lubricating oils can be analyzed by measuring the content of antioxidants in lubricating oils by infrared spectroscopy.

[0064] Step S2, selecting characteristic spectral points of antioxidants and establishing a characteristic spectral point database.

[0065] Step S3, cleaning the data of the characteristic spectrum point database to remove abnormal values ​​of the data set. Analyze the characteristic band data of a certain oxidation product of the extracted lubricating oil. When judging the antioxidant performance of the oil by the content of the oxidation product of the lubricating oil, due to the machine error of the infrared spectrometer, the test method, the environmental influence, etc., there will be certain data errors. The abnormal data points with too large deviations are removed.

[0066] Step S4, initialization of training samples: Based on the preprocessed data, training samples and test samples are respectively established, and the training samples are initialized.

[0067] Step S5, setting the input matrix and the output matrix.

[0068] The input matrix and output matrix of the training sample are X and Y respectively.

[0069]

[0070] Step S6: input the training samples into the hidden layer through the input layer, and determine the radial basis center of the hidden layer.

[0071] Step S7, calculating the thresholds corresponding to the M neurons in the hidden layer, and calculating the hidden layer output through the radial basis center.

[0072] Specifically, the training samples are input into the hidden layer of the RBF neural network through the input layer, and the radial basis center C of the hidden layer is calculated: C=X'.

[0073] The number of neurons in the hidden layer is equal to the number of training samples, so the threshold corresponding to M neurons in the hidden layer is:

[0074] B1=[b 11 ,b 12 ,……b 1M ]'.

[0075]

[0076] Among them, s is the expansion speed of the radial basis function, and M is the number of neurons in the hidden layer.

[0077] The hidden layer output can be further calculated based on the radial basis center:

[0078] A=[a1,a2,……a m ].

[0079]

[0080] Where Xi is the vector of the i-th training sample.

[0081] Step S8, calculating the hidden layer output, wherein the hidden layer output is the output layer input.

[0082] Step S9, calculate the connection weight and output layer threshold. Calculate the connection weight W and output layer threshold B2:

[0083]

[0084] B2=[b 21 ,b 22,……b 2N ]'.

[0085] [WB2]·[AI]=Y.

[0086] Among them, w ij represents the connection weight between the jth hidden neuron and the ith output neuron, N is the number of output layers, I = [1,1…1] 1×M . Where i, j, M, and N are positive integers.

[0087] Step S10, output the result. The output layer input variables are outputted through the purelin function linear weighted sum, and then the neural network training is terminated and the result predicted by the training sample is calculated. The effectiveness and accuracy of the method can be further evaluated based on the validation sample.

[0088] This patent proposes a method for studying the antioxidant properties of fan lubricating oil based on RBF neural network. Under the premise of knowing the type of antioxidant, the RBF neural network algorithm is used to measure the content of a certain oxidation product in the lubricating oil in use to evaluate the antioxidant performance of the lubricating oil.

[0089] By analyzing the changes in the content of oxidation products generated during the oxidation process of the lubricating oil in use and establishing a functional relationship between the content of lubricating oil oxidation products and the anti-oxidation property of the lubricating oil, it is possible to quantitatively judge the changes in the anti-oxidation performance of the oil and effectively understand the health status of the lubricating oil of the wind turbine.

[0090] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0091] In this embodiment, a device for obtaining the amount of generated electricity is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be repeated here. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0092] Figure 4: is a structural block diagram of a device for obtaining power generation according to an embodiment of the present application, the device comprising:

[0093] An obtaining module 42 is used to analyze the product content of the oxidation products of the fan lubricating oil configured for the power generation fan by using a neural network model to obtain the antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained by using the historical product content of the historical oxidation products generated when the power generation fan uses the historical fan lubricating oil as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample;

[0094] A determination module 44 is used to determine, for a plurality of fan lubricating oils, a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils;

[0095] The acquisition module 46 is used to configure the target wind turbine lubricant on the power-generating wind turbine and acquire the amount of power generated when the power-generating wind turbine uses the target wind turbine lubricant to operate.

[0096] Through the above-mentioned device, the product content of the oxidation product of the fan lubricant configured for the power generation wind turbine is analyzed by using a neural network model to obtain the antioxidant parameter value corresponding to the fan lubricant, wherein the neural network model is trained using the historical product content of the historical oxidation product generated when the power generation wind turbine uses the historical fan lubricant as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricant as an output sample; for multiple fan lubricants, a target fan lubricant with the maximum antioxidant parameter value is determined from the multiple fan lubricants; the target fan lubricant is configured on the power generation wind turbine, and the power generation generated when the power generation wind turbine uses the target fan lubricant is obtained. Under the premise of knowing the type of antioxidant, the content of the oxidation product can be analyzed based on the neural network model to evaluate the antioxidant performance of the lubricant, and a fan lubricant with high antioxidant performance can be selected for the power generation wind turbine, thereby solving the problem of insufficient power generation due to the use of lubricants with insufficient antioxidant properties for power generation in the power generation wind turbine, improving the power generation efficiency, and thereby increasing the power generation.

[0097] In an exemplary embodiment, the neural network model at least includes a radial basis function neural network model, and the above-mentioned module 42 is also used to sequentially set an input layer, a hidden layer and an output layer in the radial basis function neural network model; the product content of the oxidation product is input into the input layer, and the first feature output by the input layer is input into the hidden layer, so as to use the hidden layer to perform a nonlinear transformation on the first feature using a radial basis function to obtain a second feature; the second feature is linearly weighted using the output layer to output the antioxidant parameter value corresponding to the fan lubricating oil.

[0098] In an exemplary embodiment, the obtaining module 42 is further used to determine the radial basis center of the hidden layer, calculate the threshold corresponding to the M neurons in the hidden layer, and calculate the second feature through the radial basis center;

[0099] The radial basis center of the hidden layer is expressed as: C=X', X is the input matrix corresponding to the first feature,

[0100] In an exemplary embodiment, the obtaining module 42 is also used to obtain the connection weight between the hidden layer and the output layer, and to set an output layer threshold for the output layer; the antioxidant parameter value is calculated using the connection weight, the output layer threshold and the first feature.

[0101] In an exemplary embodiment, the obtaining module 42 is further used to represent the matrix Y of the antioxidant parameter values ​​by the following formula:

[0102] [WB2]·[AI]=Y;

[0103] Wherein, W is the connection weight, w ij represents the connection weight between the jth hidden neuron and the ith output neuron, B2 is the output layer threshold, B2 = [b 21 ,b 22 ,……b 2N ]', N is the number of output layers, A is the first feature, I = [1,1…1] 1×M .

[0104] In an exemplary embodiment, the device for acquiring the power generation amount also includes a testing module, which is used to perform an oxidation test on the fan lubricating oil configured for the power generation fan using an infrared spectrometer before using a neural network model to analyze the product content of the oxidation products of the fan lubricating oil configured for the power generation fan, so as to obtain oxidation band data of the fan lubricating oil, wherein the oxidation band data at least includes characteristic band data of different oxidation products generated by the fan lubricating oil when an oxidation reaction occurs; perform data cleaning on the oxidation band data to obtain clean data after cleaning; and determine the product content of the different oxidation products from the clean data.

[0105] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0106] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0107] S1, using a neural network model to analyze the product content of oxidation products of the fan lubricating oil configured for the power generation fan to obtain the antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using the historical product content of the historical oxidation products generated when the power generation fan uses the historical fan lubricating oil as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample;

[0108] S2, for a plurality of fan lubricating oils, determining a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils;

[0109] S3, configuring the target wind turbine lubricant oil on the power-generating wind turbine, and obtaining the amount of power generated when the power-generating wind turbine uses the target wind turbine lubricant oil to work.

[0110] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0111] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0112] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0113] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0114] S1, using a neural network model to analyze the product content of oxidation products of the fan lubricating oil configured for the power generation fan to obtain the antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using the historical product content of the historical oxidation products generated when the power generation fan uses the historical fan lubricating oil as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample;

[0115] S2, for a plurality of fan lubricating oils, determining a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils;

[0116] S3, configuring the target wind turbine lubricant oil on the power-generating wind turbine, and obtaining the amount of power generated when the power-generating wind turbine uses the target wind turbine lubricant oil to work.

[0117] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0118] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0119] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0120] An embodiment of the present application also provides a computer program, which includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in any one of the above method embodiments.

[0121] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0122] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0123] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for obtaining power generation, characterized in that: include: The product content of oxidation products of the fan lubricating oil configured for the power-generating fan is analyzed by using a neural network model to obtain an antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained using the historical product content of the historical oxidation products generated when the power-generating fan uses the historical fan lubricating oil as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample; for a plurality of fan lubricating oils, determining a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils; The target wind turbine lubricant is configured on the power-generating wind turbine, and the amount of power generated when the power-generating wind turbine uses the target wind turbine lubricant to operate is obtained.

2. The method according to claim 1, characterized in that The neural network model at least includes a radial basis function neural network model, and obtains the anti-oxidation parameter value corresponding to the fan lubricating oil. The neural network model is used to analyze the product content of the oxidation product of the fan lubricating oil configured for the power generation fan, including: In the radial basis function neural network model, an input layer, a hidden layer and an output layer are sequentially set; Inputting the product content of the oxidation product into the input layer, inputting the first feature output by the input layer into the hidden layer, and using the hidden layer to perform nonlinear transformation on the first feature using a radial basis function to obtain a second feature; The output layer is used to perform a linear weighted operation on the second feature, and an anti-oxidation parameter value corresponding to the fan lubricating oil is output.

3. The method according to claim 2, characterized in that Inputting the first feature output by the input layer into the hidden layer, so as to use the hidden layer to perform nonlinear transformation on the first feature using a radial basis function to obtain a second feature, comprising: Determine the radial basis center of the hidden layer, calculate the thresholds corresponding to the M neurons in the hidden layer, and calculate the second feature through the radial basis center; The radial basis center of the hidden layer is expressed as: C=X', X is the input matrix corresponding to the first feature, 4. The method according to claim 2, characterized in that: Using the output layer to perform a linear weighted operation on the second feature, and outputting an anti-oxidation parameter value corresponding to the fan lubricating oil, including: Obtaining a connection weight between the hidden layer and the output layer, and setting an output layer threshold for the output layer; The antioxidant parameter value is calculated using the connection weights, the output layer threshold and the first feature.

5. The method according to claim 4, characterized in that The matrix Y of the antioxidant parameter values ​​is represented by the following formula: [WB2]·[AI]=Y; Wherein, W is the connection weight, w ij represents the connection weight between the jth hidden neuron and the ith output neuron, B2 is the output layer threshold, B2=[b 21 ,b 22 ,……b 2N ]', N is the number of output layers, A is the first feature, I = [1,1 … 1] 1×M .

6. The method according to claim 1, characterized in that Before using the neural network model to analyze the product content of the oxidation products of the fan lubricating oil configured for the power generation fan, the method further includes: using an infrared spectrometer to perform an oxidation test on the fan lubricating oil configured for the power generation fan to obtain oxidation band data of the fan lubricating oil, wherein the oxidation band data at least includes characteristic band data of different oxidation products generated when the fan lubricating oil undergoes an oxidation reaction; Cleaning the oxidation band data to obtain clean data; Product levels of the different oxidation products are determined from the clean data.

7. A device for obtaining power generation, characterized in that: include: A module is provided for analyzing the product content of oxidation products of the fan lubricating oil configured for the power generation fan by using a neural network model to obtain the antioxidant parameter value corresponding to the fan lubricating oil, wherein the neural network model is trained by using the historical product content of the historical oxidation products generated when the power generation fan uses the historical fan lubricating oil as an input sample and the historical antioxidant parameter value corresponding to the historical fan lubricating oil as an output sample; a determination module, configured to determine, for a plurality of fan lubricating oils, a target fan lubricating oil having a maximum anti-oxidation parameter value from the plurality of fan lubricating oils; The acquisition module is used to configure the target wind turbine lubricant on the power-generating wind turbine and acquire the amount of power generated when the power-generating wind turbine uses the target wind turbine lubricant to work.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images