A power transmission and transformation project investment prediction method and device

Abstract

The application provides a power transmission and transformation project investment prediction method and device, and the method comprises the following steps: acquiring a data set of a power transmission and transformation project; processing the data set to obtain a feature set; inputting the feature set into a power transmission and transformation project investment prediction model to perform investment prediction, and obtaining power transmission and transformation project investment prediction data; wherein the power transmission and transformation project investment prediction model is obtained by inputting a training data set into a preset neural network model for training, and an adjustment amount of a preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function. The scheme improves the efficiency of power transmission and transformation project investment prediction, and improves the stability of power transmission and transformation project investment prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission and transformation project investment prediction, in particular to a power transmission and transformation project investment prediction method and device. BACKGROUND

[0002] With the acceleration of the development speed of domestic power grid projects, the investment scale of the power industry is rapidly growing, and cost analysis is an indispensable part of power grid project management and construction, and is the basis for analyzing the investment benefit of power grid projects, directly affecting the income and management level of power grid enterprises. Due to the complexity of the construction process of power transmission and transformation projects, the long construction period, large investment amount, and great influence of environmental factors, the prediction of the cost of power transmission and transformation projects is greatly improved.

[0003] In the prior art, the current power transmission and transformation project investment prediction method is prone to local minimum value, cannot guarantee the global optimality of the network each time, and the selection of the network iteration step and the inertia coefficient is determined by experience, and improper selection can cause network oscillation and even lead to network convergence, affecting the prediction stability; at the same time, the convergence speed is slow and the prediction efficiency is low. SUMMARY

[0004] The present application provides a power transmission and transformation project investment prediction method and device to solve the problem of unsatisfactory power transmission and transformation project investment prediction effect.

[0005] To solve the above technical problems, the technical solutions of the present application are as follows:

[0006] A power transmission and transformation project investment prediction method, comprising:

[0007] Obtaining a data set of a power transmission and transformation project;

[0008] Processing the data set to obtain a feature set;

[0009] Inputting the feature set into a power transmission and transformation project investment prediction model for investment prediction to obtain power transmission and transformation project investment prediction data; wherein the power transmission and transformation project investment prediction model is obtained by inputting a training data set into a preset neural network model and training, and the adjustment amount of a preset parameter optimizer of the preset neural network model is obtained by a preset adjustment function.

[0010] Optionally, processing the data set to obtain a feature set comprises:

[0011] According to the data set, obtaining the maximum and minimum values in the data group of the data set;

[0012] According to the formula: y = (x-min Value) / (max Value-min Value), all data points in the data group of the data set are normalized to obtain a feature set;

[0013] Wherein, x is the data of the data point before normalization, y is the data after normalization, max Value and min Value correspond to the maximum value and the minimum value in the data group of the data point respectively.

[0014] Optionally, the training process of the power transmission and transformation project investment prediction model comprises:

[0015] Obtaining a training data set;

[0016] Processing the training data set to obtain a training feature set;

[0017] Inputting the training feature set into an input layer of the preset neural network model to obtain a first layer output;

[0018] Inputting the first layer output and a first target parameter into a hidden layer to obtain a second layer output;

[0019] Inputting the second layer output and a second target parameter into an output layer to obtain a power transmission and transformation project investment prediction data determination model.

[0020] Optionally, inputting the training feature set into an input layer of the preset neural network model to obtain a first layer output; inputting the first layer output and a first target parameter into a hidden layer to obtain a second layer output; inputting the second layer output and a second target parameter into an output layer to obtain a power transmission and transformation project investment prediction data determination model, comprising:

[0021] As a transfer function:

[0022]

[0023]

[0024] The output of the input layer node I is: Equal to the input x of the input layer node I I ;

[0025] The input of each node of the hidden layer is:

[0026]

[0027] The output of each node of the hidden layer is:

[0028]

[0029] ​wherein, W JI is the connection weight between the hidden layer node J and the input layer node, and J is the bias of the hidden layer node J;

[0030] The input of the output layer node is:

[0031]

[0032] The output of the output layer node is:

[0033]

[0034] wherein, W LJ is the connection weight between the output layer node L and the hidden layer node J, and L is the bias of the output layer node L.

[0035] Optionally, the power transmission and transformation project investment prediction method further comprises:

[0036] The prediction accuracy of the power transmission and transformation project investment prediction model is evaluated through a mean square error function:

[0037]

[0038] The prediction accuracy of the power transmission and transformation project investment prediction model is evaluated through a mean square error function:

[0039] wherein, P is the training sample number, t PI is the target output of the Ith output unit of the Pth sample, and O PI is the network operation result of the Ith output unit of the Pth sample.

[0040] Optionally, the power transmission and transformation project investment prediction method further comprises:

[0041] The first target parameter and the second target parameter are updated through a preset parameter optimizer and a preset training set:

[0042] The updating of the second target parameter is according to the connection weight W LJ between the output layer and the hidden layer, and specifically:

[0043]

[0044]

[0045] The updating of the first target parameter is according to the connection weight W JI between the hidden layer and the input layer, and specifically:

[0046]

[0047]

[0048] wherein t L is the target output of the Lth output unit, is the network operation result of the Lth output unit, K is the iteration number, η is the learning rate, i.e., the gain coefficient of the weight value, and α is the momentum term for adjusting the convergence speed of learning, 0<η<1 and 0<α<1.

[0049] Optionally, the adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function, and the adjustment amount comprises:

[0050] The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function:

[0051] dW(i,j)=mc*dW prev (i,j)+mc*lr*gW(i,j) / dW(i,j), wherein dW is the adjustment amount of the network connection weight, gW is the gradient vector of the network performance with respect to the network connection weight, mc is the momentum factor, 0.9<mc<1, lr is the learning efficiency, and lr is 0.001, 0.0001, 0.00001 or 0.000001. LJ (K)-W LJ (K-1)] and [W JI (K)-W JI (K-1)].

[0052] wherein dW is the adjustment amount of the network connection weight, gW is the gradient vector of the network performance with respect to the network connection weight, mc is the momentum factor, 0.9<mc<1, lr is the learning efficiency, and lr is 0.001, 0.0001, 0.00001 or 0.000001.

[0053] The application further provides a power transmission and transformation project investment prediction device, comprising:

[0054] An acquisition module is configured to acquire a data set of the power transmission and transformation project.

[0055] A processing module is configured to process the data set to obtain a feature set, input the feature set into a power transmission and transformation project investment prediction model to perform investment prediction, obtain power transmission and transformation project investment prediction data, and output the power transmission and transformation project investment prediction data.

[0056] The application further provides a computing device, comprising a processor and a memory storing a computer program, wherein the computer program is executed by the processor to perform the method described above.

[0057] The application further provides a computer readable storage medium storing instructions, wherein the instructions are executed on a computer to cause the computer to perform the method described above.

[0058] The above-mentioned scheme of the application has at least the following beneficial effects:

[0059] The scheme of the present application comprises: obtaining a data set of a power transmission and transformation project; processing the data set to obtain a feature set; inputting the feature set into a power transmission and transformation project investment prediction model to perform investment prediction and obtain power transmission and transformation project investment prediction data; wherein the power transmission and transformation project investment prediction model is obtained by inputting a training data set into a preset neural network model for training, and an adjustment amount of a preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function. The scheme of the present application improves the efficiency of power transmission and transformation project investment prediction, and improves the stability of power transmission and transformation project investment prediction. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a flowchart of a power transmission and transformation project investment prediction method provided by an embodiment of the present application;

[0061] Figure 2 is a neural network structure diagram of a power transmission and transformation project investment prediction model provided by an embodiment of the present application;

[0062] Figure 3 is a training flowchart of a power transmission and transformation project investment prediction model provided by an embodiment of the present application;

[0063] Figure 4 is a module diagram of a power transmission and transformation project investment prediction device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0064] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present application can be more thoroughly understood and so that the scope of the present application can be completely conveyed to those skilled in the art.

[0065] As shown in Figure 1 , an embodiment of the present application proposes a power transmission and transformation project investment prediction method, comprising:

[0066] Step 11, obtaining a data set of a power transmission and transformation project;

[0067] Step 12, processing the data set to obtain a feature set;

[0068] Step 13, inputting the feature set into a power transmission and transformation project investment prediction model to perform investment prediction and obtain power transmission and transformation project investment prediction data, and outputting;

[0069] The power transmission and transformation project investment prediction model is obtained by inputting a training data set into a preset neural network model for training, and an adjustment amount of a preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function.

[0070] In this embodiment, a power transmission and transformation project investment prediction model is constructed by optimizing the neural network model through a preset adjustment function. The acquired power transmission and transformation project dataset is processed to obtain a feature set, which is then input into the power transmission and transformation project investment prediction model for investment prediction. This improves both the efficiency and stability of power transmission and transformation project investment prediction. It also solves the problems in existing technologies where power transmission and transformation project investment prediction methods are prone to getting trapped in local minima, cannot guarantee the global optimality of the network in each training iteration, and suffer from prediction instability caused by improper selection of network iteration step size and inertia coefficient, as well as low prediction efficiency.

[0071] In an optional embodiment of the present invention, step 12 includes:

[0072] Step 121: Based on the dataset, obtain the maximum and minimum values ​​in the data groups of the dataset;

[0073] Step 122: According to the formula: y = (x - min Value) / (max Value - min Value), normalize all data points in the dataset to obtain the feature set;

[0074] Where x is the data point before normalization, y is the data after normalization, and maxValue and minValue correspond to the maximum and minimum values ​​in this set of data points, respectively.

[0075] In this embodiment, the power transmission and transformation project cost dataset is a structured dataset containing many data items, such as equipment price, equipment quantity, cable price, cable quantity, substation area, concrete usage, line length, conductor cross-section, conductor price, tower material usage, tower material price, earthwork volume, and topography. The dataset is stored in tabular form, with each column representing a specific data type. Furthermore, the power engineering cost dataset is large in volume and contains various data types, including numbers and text, requiring large-scale data analysis and processing techniques. Since power transmission and transformation project construction involves multiple factors and a highly complex process, the power transmission and transformation project cost dataset contains a wealth of information, necessitating the use of mathematical models and big data mining techniques to reduce the difficulty of data analysis and minimize errors. Finally, because power engineering cost datasets typically contain important market data and confidential information, feasible data processing and analysis must be conducted while ensuring data security.

[0076] The maximum value and the minimum value in the data set are obtained according to the data set, all data points in the data set are normalized according to the formula: y=(x-min Value) / (max Value-min Value), and a feature set is obtained, so that the data in the feature set are in a uniform scale in the range of [0, 1], the influence of different scales on data analysis is eliminated, the convergence speed of a prediction algorithm is improved, and the rapidity, accuracy and reliability of prediction are improved.

[0077] In an optional embodiment of the present application, the training process of the power transmission and transformation project investment prediction model comprises:

[0078] In step 131, a training data set is obtained.

[0079] In step 132, the training data set is processed to obtain a training feature set.

[0080] In step 133, the training feature set is input into an input layer of the preset neural network model for processing to obtain a first layer output.

[0081] In step 134, the first layer output and a first target parameter are input into a hidden layer for processing to obtain a second layer output.

[0082] In step 135, the second layer output and a second target parameter are input into an output layer for processing to obtain a power transmission and transformation project investment prediction data determination model.

[0083] In this embodiment, the settled data of power transmission and transformation projects of each voltage level in previous years is collected as the training data set.

[0084] The maximum value and the minimum value in the training data set are obtained according to the training data set, all data points in the training data set are normalized according to the formula: y=(x-min Value) / (max Value-min Value), and a training feature set is obtained.

[0085] The training feature set is input into an input layer of the preset neural network model for processing to obtain a first layer output, the first layer output and a first target parameter are input into a hidden layer for processing to obtain a second layer output, and the second layer output and a second target parameter are input into an output layer for processing to obtain a power transmission and transformation project investment prediction data determination model.

[0086] The function is used as a transfer function.

[0087]

[0088] ​

[0089] The output of the first layer is specifically the output of the input layer node I equal to the input x of the input layer node I I ;

[0090] The input of each node of the hidden layer is

[0091]

[0092] The output of the second layer is specifically the output of each node of the hidden layer

[0093]

[0094] wherein, W JI is the connection weight between the hidden layer node J and the input layer node, and θ J is the bias of the hidden layer node J

[0095] The input of the output layer node is

[0096]

[0097] The investment prediction data of the power transmission and transformation project is specifically the output of the output layer node

[0098]

[0099] wherein, W LJ is the connection weight between the output layer node L and the hidden layer node J, and θ L is the bias of the output layer node L

[0100] In an optional embodiment of the present application, the power transmission and transformation project investment prediction method further comprises

[0101] Step 14, the prediction accuracy of the power transmission and transformation project investment prediction model is evaluated by the mean square error function

[0102]

[0103]

[0104] wherein, P is the training sample number, t PI is the target output of the Ith output unit of the Pth sample, and O PI is the network operation result of the Ith output unit of the Pth sample

[0105] In this embodiment, for a given training sample set (x p1 , x p2 ,...x pn )→(t p1 , t p2 ,...t pn ​), p = 1, 2,..., p is a sample number;

[0106] According to the network operation result of the output unit, the obtained power transmission and transformation project investment prediction data, and the target output of the training sample set, a mean square error is obtained through a mean square error function to verify the prediction accuracy of the power transmission and transformation project investment prediction model; if the calculated mean square error is large, the first target parameter and the second target parameter are updated, and the power transmission and transformation project investment prediction model is trained again until the mean square error is minimized, and the training is completed.

[0107] In an optional embodiment of the present application, the power transmission and transformation project investment prediction method further comprises:

[0108] Step 15, updating the first target parameter and the second target parameter through the preset parameter optimizer and the preset training set:

[0109] The updating of the second target parameter is according to the connection weight W LJ between the output layer and the hidden layer, and specifically:

[0110]

[0111]

[0112] The updating of the first target parameter is according to the connection weight W JI between the hidden layer and the input layer, and specifically:

[0113]

[0114]

[0115] wherein t L is the target output of the Lth output unit, is the network operation result of the Lth output unit, K is the iteration number, η is the learning rate, that is, the gain coefficient of the weight value, and α is the momentum term for adjusting the convergence speed of learning, 0 < η < 1, and 0 < α < 1.

[0116] In this embodiment, the second target parameter, that is, the connection weight between the output layer and the hidden layer, and the first target parameter, that is, the connection weight between the hidden layer and the input layer, are updated through the preset parameter optimizer and the preset training set and according to the gradient direction of the mean square error, so that the mean square error is minimized, and the optimization of the power transmission and transformation project investment prediction model is realized.

[0117] In an optional embodiment of the present application, the adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function, which comprises:

[0118] Through a preset adjustment function:

[0119] dW(i,j) = mc * dW prev (i,j) + mc * lr * gW(i,j) / dW(i,j) to obtain the adjustment amount of the preset parameter optimizer of the preset neural network model, and the adjustment amount corresponds to [W LJ (K) - W LJ (K - 1)] and [W JI (K) - W JI (K - 1)];

[0120] where dW is the adjustment amount of the network connection weight, gW is the gradient vector of the network performance with respect to the network connection weight, mc is the momentum factor, 0.9 < mc < 1, lr is the learning efficiency, and lr is 0.001, 0.0001, 0.00001, or 0.000001.

[0121] In this embodiment, specifically, mc is 0.95 and lr is 0.00001. The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through the preset adjustment function, that is, the negative gradient descent algorithm of the BP neural network is improved, and the adaptive learning rate momentum gradient descent backpropagation algorithm is applied to adjust the adjustment amount of the preset parameter optimizer. The learning rate is adaptively adjusted according to the network performance, thereby improving the network training preparation efficiency, increasing the training speed of the network, and making the network have better stability.

[0122] Such as Figure 2 and Figure 3 shown, collect the settled data of transmission and transformation projects of each voltage level in the past 5 years,剔除 special projects and error factors to obtain a dataset, where the training set is 60%, the validation set is 20%, and the test set is 20%;

[0123] Normalize the training dataset through Origin2022 to obtain a training feature set;

[0124] Input the training feature set into the preset neural network model, that is, input the given input vector and the target output;

[0125] The preset neural network model obtains the outputs of each unit in the hidden layer and the output layer according to the activation function;

[0126] Obtain the deviation according to the outputs of each unit and the target output, that is, the mean square error;

[0127] Judge the mean square error. When the mean square error is small and meets the requirements, the training is completed;

[0128] When the mean square error is large and does not meet the requirements, obtain the adjustment amount of the preset parameter optimizer according to the preset adjustment function; It should be noted that there is an unclear expression "剔除 special projects and error factors" in the original text. You may need to check and correct it according to the actual situation.

[0129] The preset parameter optimizer and the preset training set are used to update the connection weights between the output layer and the hidden layer and the connection weights between the hidden layer and the input layer, and the training is performed again until the requirements are met, so that an ideal power transmission and transformation project investment prediction model is obtained.

[0130] As shown in Figure 4 The embodiment of the present application also provides a power transmission and transformation project investment prediction device 20, which comprises:

[0131] An acquisition module 21 is configured to acquire radial magnetic flux leakage detection data and axial magnetic flux leakage detection data of a pipeline.

[0132] A processing module 22 is configured to perform feature map processing on the radial magnetic flux leakage detection data and the axial magnetic flux leakage detection data to obtain feature quantities, input the feature quantities into a pipeline magnetic flux leakage defect detection model to perform magnetic flux leakage defect detection, obtain pipeline magnetic flux leakage defect data, and output the pipeline magnetic flux leakage defect data.

[0133] Optionally, the data set is processed to obtain a feature set, including:

[0134] According to the data set, the maximum value and the minimum value in the data group of the data set are obtained.

[0135] According to a formula: y=(x-min Value) / (max Value-min Value), all data points in the data group of the data set are normalized to obtain a feature set.

[0136] Wherein, x is the data of the data point before normalization, y is the data after normalization, max Value and min Value correspond to the maximum value and the minimum value in the data group of the data point, respectively.

[0137] Optionally, the training process of the power transmission and transformation project investment prediction model comprises:

[0138] A training data set is acquired.

[0139] The training data set is processed to obtain a training feature set.

[0140] The training feature set is input into an input layer of the preset neural network model to obtain a first layer output.

[0141] The first layer output and a first target parameter are input into a hidden layer to obtain a second layer output.

[0142] The second layer output and a second target parameter are input into an output layer to obtain a power transmission and transformation project investment prediction data determination model.

[0143] Optionally, the training feature set is input into an input layer of the preset neural network model for processing to obtain a first layer output; the first layer output and a first target parameter are input into a hidden layer for processing to obtain a second layer output; and the second layer output and a second target parameter are input into an output layer for processing to obtain a power transmission and transformation project investment prediction data determination model, comprising:

[0144] The function is:

[0145]

[0146] as a transfer function;

[0147] The output of the input layer node I is: equal to the input x of the input layer node I I ;

[0148] The input of each node of the hidden layer is:

[0149]

[0150] The output of each node of the hidden layer is:

[0151]

[0152] wherein, W JI is the connection weight between the hidden layer node J and the input layer node, and θ J is the bias of the hidden layer node J;

[0153] The input of the output layer node is:

[0154]

[0155] The output of the output layer node is:

[0156]

[0157] wherein, W LJ is the connection weight between the output layer node L and the hidden layer node J, and θ L is the bias of the output layer node L.

[0158] Optionally, the power transmission and transformation project investment prediction method further comprises:

[0159] The prediction accuracy of the power transmission and transformation project investment prediction model is evaluated through a mean square error function:

[0160]

[0161]

[0162] wherein, P is the training sample number, t PI ​O PI is the network operation result of the Ith output unit of the Pth sample.

[0163] Optionally, the power transmission and transformation project investment prediction method further comprises:

[0164] The first target parameter and the second target parameter are updated through the preset parameter optimizer and the preset training set:

[0165] The second target parameter is updated according to the connection weight W between the output layer and the hidden layer. LJ Specifically, the second target parameter is updated according to the connection weight W between the output layer and the hidden layer.

[0166]

[0167]

[0168] The first target parameter is updated according to the connection weight W between the hidden layer and the input layer. JI Specifically, the first target parameter is updated according to the connection weight W between the hidden layer and the input layer.

[0169]

[0170]

[0171] wherein t L is the target output of the Lth output unit, is the network operation result of the Lth output unit, K is the iteration number, η is the learning rate, i.e., the gain coefficient of the weight value, α is the momentum term, used for adjusting the convergence speed of learning, 0<η<1, 0<α<1.

[0172] Optionally, the adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function, comprising:

[0173] The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function:

[0174] dW(i,j)=mc*dW prev (i,j)+mc*lr*gW(i,j) / dW(i,j) The adjustment amount of the preset parameter optimizer of the preset neural network model corresponds to [W LJ (K)-W LJ (K-1)] and [W JI (K)-W JI (K-1)].

[0175] wherein dW is an adjustment amount of the network connection weight, gW is a gradient vector of the network performance to the network connection weight, mc is a momentum factor, 0.9 < mc < 1, lr is a learning efficiency, and lr is 0.001, 0.0001, 0.00001, or 0.000001.

[0176] It should be noted that the device corresponds to the above method, and all implementation manners in the above method embodiments are applicable to the device embodiments and can achieve the same technical effects.

[0177] In the embodiments of the present application, a computing device is also provided, which comprises a processor and a memory storing a computer program, and the computer program is executed by the processor to perform the method described in the above embodiments. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0178] In the embodiments of the present application, a computer readable storage medium is also provided, which stores instructions, and when the instructions are executed on a computer, the computer performs the method described in the above embodiments. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0179] Those skilled in the art can understand that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0180] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the above method embodiments, which will not be repeated here.

[0181] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0182] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0183] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0184] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk, and various program code storage media.

[0185] In addition, it should be noted that in the device and method of the present application, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombination should be considered as equivalent solutions of the present application. Moreover, the steps of executing the above series of processes can naturally be executed in time sequence according to the order of description, but do not necessarily have to be executed in time sequence, and some steps can be executed in parallel or independently of each other. It can be understood by those skilled in the art that all or any steps or components of the method and device of the present application can be realized in hardware, firmware, software or their combination in any computing device (including processor, storage medium, etc.) or network of computing devices, which can be realized by those skilled in the art using their basic programming skills after reading the description of the present application.

[0186] Therefore, the object of the present application can also be achieved by running a program or a set of programs on any computing device. The computing device can be a commonly known general purpose device. Therefore, the object of the present application can also be achieved only by providing a program product containing program code for implementing the method or device. That is, such a program product also constitutes the present application, and a storage medium storing such a program product also constitutes the present application. Obviously, the storage medium can be any commonly known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present application, obviously, the components or steps can be decomposed and / or recombined. These decompositions and / or recombination should be considered as equivalent solutions of the present application. Moreover, the steps of performing the above series of processes can naturally be executed in time sequence according to the order of description, but do not necessarily have to be executed in time sequence. Some steps can be executed in parallel or independently of each other.

[0187] The in-orbit data transmission method, system and device described in the above embodiments of the present application are applied to the transmission of physiological parameters of astronauts in orbit. The satellite communication technology is combined with the physiological telemetry parameters of long-term astronauts in orbit. Real-time data transmission and monitoring are realized by using satellite communication transmission technology. The stable transmission of data is ensured, the risk of data transmission error and loss is reduced, the integrity and accuracy of the physiological parameter data of astronauts are ensured, the health status of astronauts can be monitored and evaluated in real time by the medical team on the ground, and feedback and decisions can be made in time to ensure the health and safety of astronauts in orbit.

[0188] The above is the preferred embodiment of the present application. It should be noted that for ordinary skilled persons in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered within the scope of protection of the present application.

Claims

1. A method for analyzing power transmission and transformation cost data, characterized in that, include: Obtain a dataset of power transmission and transformation costs, which includes: number of equipment, number of cables, substation area, amount of concrete used, line length, conductor cross-section, tower material used, earthwork volume, and terrain. The dataset is processed to obtain a feature set; The feature set is input into the power transmission and transformation cost data analysis model for analysis to obtain power transmission and transformation cost data analysis data; wherein, the power transmission and transformation cost data analysis model is obtained by inputting the training dataset into a preset neural network model for training, and the adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function; The training process of the power transmission and transformation cost data analysis model includes: Obtain the training dataset; The training dataset is processed to obtain a training feature set; The training feature set is input into the input layer of the preset neural network model for processing to obtain the first layer output; The first layer output and the first target parameter are input into the hidden layer for processing to obtain the second layer output. The second layer output and the second target parameter are input into the output layer for processing to obtain the power transmission and transformation cost data analysis data determination model. The training feature set is input into the input layer of the preset neural network model for processing to obtain the first layer output; the first layer output and the first target parameter are input into the hidden layer for processing to obtain the second layer output; the second layer output and the second target parameter are input into the output layer for processing to obtain the power transmission and transformation cost data analysis and determination model, including: With function: As a transfer function; The output of input layer node I is: , equal to the input of input layer node I ; The inputs to each node in the hidden layer are: The outputs of each node in the hidden layer are: in, W JI Let J be the connection weight between the hidden layer node J and the input layer node. θ J The bias of hidden layer node J; The input to the output layer node is: The output of the output layer node is: in, W LJ The connection weights between output layer node L and hidden layer node J. θ L The bias of the output layer node L; The methods for analyzing power transmission and transformation cost data also include: Through the mean square error function: Evaluate the predictive accuracy of the power transmission and transformation cost data analysis model; Where P is the training sample number. t PI The target output of the i-th output unit of the p-th sample. O PI The network computation result of the i-th output unit of the p-th sample; The methods for analyzing power transmission and transformation cost data also include: The first and second target parameters are updated using a preset parameter optimizer and a preset training set: The second objective parameter is updated based on the connection weights between the output layer and the hidden layer. W LJ Specifically: The first objective parameter is updated based on the connection weights between the hidden layer and the input layer. W JI Specifically: in The target output of the Lth output unit is... The result of the network operation for the Lth output unit is given, and K is the number of iterations. The learning rate is the gain coefficient of the weights. This is the momentum term, used to adjust the convergence speed of the learning process. , ; The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function, including: By using a preset adjustment function: The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained, and the adjustment amount corresponds to and ; in This is an adjustment amount for network connection rights. Let be the gradient vector of network performance with respect to network connection weights. Momentum factor , For learning efficiency, The value is 0.00001.

2. The method for analyzing power transmission and transformation cost data according to claim 1, characterized in that, The dataset is processed to obtain a feature set, including: Based on the dataset, obtain the maximum and minimum values ​​in the data groups of the dataset; According to the formula: y = (x-min Value) / (max Value-min Value), all data points in the dataset are normalized to obtain the feature set; Where x is the data point before normalization, y is the data after normalization, and max value and min value correspond to the maximum and minimum values ​​in this set of data points, respectively.

3. A power transmission and transformation cost data analysis device, characterized in that, include: The acquisition module is used to acquire the dataset of the power transmission and transformation project. The dataset includes: number of equipment, number of cables, substation area, amount of concrete used, line length, conductor cross-section, tower material used, earthwork volume, and terrain. The processing module is used to process the dataset to obtain a feature set; input the feature set into the power transmission and transformation cost data analysis model for analysis to obtain power transmission and transformation cost data analysis data, and output the result. The training process of the power transmission and transformation cost data analysis model includes: Obtain the training dataset; The training dataset is processed to obtain a training feature set; The training feature set is input into the input layer of a preset neural network model for processing to obtain the first layer output; The first layer output and the first target parameter are input into the hidden layer for processing to obtain the second layer output. The second layer output and the second target parameter are input into the output layer for processing to obtain the power transmission and transformation cost data analysis data determination model. The training feature set is input into the input layer of the preset neural network model for processing to obtain the first layer output; the first layer output and the first target parameter are input into the hidden layer for processing to obtain the second layer output; the second layer output and the second target parameter are input into the output layer for processing to obtain the power transmission and transformation cost data analysis and determination model, including: With function: As a transfer function; The output of input layer node I is: , equal to the input of input layer node I ; The inputs to each node in the hidden layer are: The outputs of each node in the hidden layer are: in, W JI Let J be the connection weight between the hidden layer node J and the input layer node. θ J The bias of hidden layer node J; The input to the output layer node is: The output of the output layer node is: in, W LJ The connection weights between output layer node L and hidden layer node J. θ L The bias of the output layer node L; The processing module is also used for: Through the mean square error function: Evaluate the predictive accuracy of the power transmission and transformation cost data analysis model; Where P is the training sample number. t PI The target output of the i-th output unit of the p-th sample. O PI The network computation result of the i-th output unit of the p-th sample; The processing module is also used for: The first and second target parameters are updated using a preset parameter optimizer and a preset training set: The second objective parameter is updated based on the connection weights between the output layer and the hidden layer. W LJ Specifically: The first objective parameter is updated based on the connection weights between the hidden layer and the input layer. W JI Specifically: in The target output of the Lth output unit is... The result of the network operation for the Lth output unit is given, and K is the number of iterations. The learning rate is the gain coefficient of the weights. This is the momentum term, used to adjust the convergence speed of the learning process. , ; The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained through a preset adjustment function, including: By using a preset adjustment function: The adjustment amount of the preset parameter optimizer of the preset neural network model is obtained, and the adjustment amount corresponds to and ; in This is an adjustment amount for network connection rights. Let be the gradient vector of network performance with respect to network connection weights. Momentum factor , For learning efficiency, The value is 0.00001.

4. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 2.

Images