An electric-hydrogen coupling conversion power grid sending project cost prediction method and terminal

Abstract

The application discloses an electric-hydrogen coupling conversion power grid sending project cost prediction method and a terminal, determines an influence factor set of the electric-hydrogen coupling conversion power grid sending project, uses a factor analysis method to extract features from the influence factor set, obtains characteristic quantities, and uses a PLS-ANN integrated prediction model to predict the cost of the electric-hydrogen coupling conversion power grid sending project according to different voltage grades based on the characteristic quantities. In this way, the factor analysis method can effectively determine the influence factors most related to the project cost, the influence factors are used as the characteristic quantities, and the PLS-ANN integrated prediction model is used to predict the project cost according to different voltage grades, so that differentiated prediction is realized, different algorithms are combined, the advantages of different algorithms are combined, and therefore, more accurate and effective project cost prediction is realized.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of engineering cost prediction, and in particular to an engineering cost prediction method for an electric-hydrogen coupling conversion power grid sending project and a terminal. BACKGROUND

[0002] For the "double carbon" target, hydrogen energy is an indispensable support carrier. Hydrogen energy is expected to make up for the shortage of electric energy and help deep decarbonization in high-energy and high-emission fields on the energy consumption side.

[0003] Reasonable prediction of the cost plays a relatively important role in the early stage of engineering construction. With the continuous improvement of the informatization level of the engineering cost management of the electric-hydrogen coupling conversion power grid sending project, more and more data related to the engineering cost are obtained. How to mine potential knowledge information in the data is the key to improving the engineering cost management efficiency. In order to provide a quantitative analysis and decision-making tool for design optimization and life cycle management of the cost, analyze historical engineering data under the same condition, and establish an effective prediction model, the reasonable control and effective management of the cost can be quantitatively supported. SUMMARY

[0004] The technical problem to be solved by the application is to provide an engineering cost prediction method for an electric-hydrogen coupling conversion power grid sending project and a terminal, which can realize more accurate and effective engineering cost prediction.

[0005] To solve the above technical problems, the technical scheme adopted by the application is as follows:

[0006] An engineering cost prediction method for an electric-hydrogen coupling conversion power grid sending project comprises the following steps:

[0007] determining an influence factor set of the engineering cost of the electric-hydrogen coupling conversion power grid sending project;

[0008] using a factor analysis method to extract features from the influence factor set to obtain characteristic quantities;

[0009] using a PLS-ANN integrated prediction model based on the characteristic quantities to predict the engineering cost of the electric-hydrogen coupling conversion power grid sending project according to different voltage grades to obtain a prediction result.

[0010] To solve the above technical problems, another technical scheme adopted by the application is as follows:

[0011] An engineering cost prediction terminal for an electric-hydrogen coupling conversion power grid sending project comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0012] determining an influence factor set of the engineering cost of the electric-hydrogen coupling conversion power grid sending project;

[0013] characteristic quantities are obtained by using factor analysis on the influence factor set;

[0014] The cost of the electricity-hydrogen coupling conversion power grid sending project is predicted based on the characteristic quantities by using a PLS-ANN integrated prediction model according to different voltage levels, and a prediction result is obtained.

[0015] The present application has the beneficial effects that the influence factor set of the engineering cost of the electricity-hydrogen coupling conversion power grid sending project is determined, the factor analysis is used to extract the characteristic quantities from the influence factor set, the PLS-ANN integrated prediction model is used based on the characteristic quantities to predict the engineering cost of the electricity-hydrogen coupling conversion power grid sending project according to different voltage levels, the most relevant influence factors are effectively determined by using the factor analysis, the characteristic quantities are used, and the PLS-ANN integrated prediction model is used to predict the engineering cost according to different voltage levels, so that the prediction is realized, different algorithms are combined, the advantages of different algorithms are combined, and more accurate and effective engineering cost prediction is realized. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A step flow chart of an electricity-hydrogen coupling conversion power grid sending project cost prediction method according to an embodiment of the present application;

[0017] Figure 2 A structural schematic diagram of an electricity-hydrogen coupling conversion power grid sending project cost prediction terminal according to an embodiment of the present application;

[0018] Figure 3 A prediction schematic diagram in an electricity-hydrogen coupling conversion power grid sending project cost prediction method according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] The technical content, purposes and effects of the present application are described in detail below by combining the embodiments with the drawings.

[0020] Please refer to Figure 1 An electricity-hydrogen coupling conversion power grid sending project cost prediction method, comprising the steps of:

[0021] determining an influence factor set of the engineering cost of the electricity-hydrogen coupling conversion power grid sending project;

[0022] characteristic quantities are obtained by using factor analysis on the influence factor set;

[0023] The cost of the electricity-hydrogen coupling conversion power grid sending project is predicted based on the characteristic quantities by using a PLS-ANN integrated prediction model according to different voltage levels, and a prediction result is obtained.

[0024] From the above description, the beneficial effects of the present application are that the influence factor set of the construction cost of the electric-hydrogen coupling conversion power grid sending-out project is determined, the factor analysis method is used to extract features from the influence factor set, the characteristic quantity is obtained, the PLS-ANN integrated prediction model is used to predict the construction cost of the electric-hydrogen coupling conversion power grid sending-out project according to the different voltage levels based on the characteristic quantity, the most relevant influence factor of the construction cost can be effectively determined by using the factor analysis method, which is used as the characteristic quantity and the PLS-ANN integrated prediction model is used to predict the construction cost according to the different voltage levels, the differentiated prediction is realized, different algorithms are combined, the advantages of different algorithms are combined, and therefore more accurate and effective construction cost prediction is realized.

[0025] Further, the electric-hydrogen coupling conversion power grid sending-out project includes an electric-hydrogen coupling conversion power grid sending-out power generation project and an electric-hydrogen coupling conversion power grid sending-out sending-out project.

[0026] The influence factor set of the construction cost of the electric-hydrogen coupling conversion power grid sending-out project includes:

[0027] The total station building area, the total station land area, the main control building area, the land area inside the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the single price of the circuit breaker, the single price of the main transformer, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line and the ground resistance are determined as the first influence factor set of the electric-hydrogen coupling conversion power grid sending-out power generation project.

[0028] The line folding length, the tower material amount, the complex terrain length, the total tower base, the wire material amount, the double circuit length, the strain angle tower base, the total earthwork amount, the foundation steel material amount, the single wire area, the single circuit length, the wire price and the foundation steel price are determined as the second influence factor set of the electric-hydrogen coupling conversion power grid sending-out sending-out project.

[0029] As can be seen from the above description, according to the engineering cost experience, the hydrogen coupling conversion power grid sending out power generation project is greatly affected by the station site geological conditions and the main equipment model, and is relatively less affected by the meteorological conditions, so the total station building area, the total station land area, the main control building land area, the land area within the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the single price of the circuit breaker, the single price of the main transformer, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line and the ground resistance are determined as the first influence factor set of the hydrogen coupling conversion power grid sending out power generation project, and the hydrogen coupling conversion power grid sending out project is affected by the path factor, the meteorological condition, the conductor model, the tower shape and the tower height four technical and other technical conditions, the line folding length, the tower material amount, the complex terrain length, the total number of poles, the wire material amount, the double circuit length, the strain angle tower base number, the total amount of earthwork, the foundation steel material amount, the single conductor area, the single circuit length, the conductor price and the foundation steel price are determined as the second influence factor set of the hydrogen coupling conversion power grid sending out project, and the preliminary confirmation of the engineering cost influence factors is realized.

[0030] Further, the influence factor set is extracted by using the factor analysis method, and the characteristic quantity includes:

[0031] The common factor variance of each influence factor in the influence factor set is calculated.

[0032] The influence factor set is sorted in descending order of the common factor variance, and a sorted influence factor set is obtained.

[0033] A preset number of influence factors are selected from the sorted influence factor set as characteristic quantities.

[0034] As can be seen from the above description, the hydrogen coupling conversion power grid sending out project cost data sample contains many influence factors, and the input data of the prediction model has the problems of high data dimension and difficult fitting, which cannot realize effective prediction of the engineering cost, so the factor analysis method is used to determine the characteristic quantity by calculating the common factor variance of each influence factor, the characteristic quantity is the characteristic data which greatly affects the prediction accuracy of the model, so as to reduce the data dimension, thereby ensuring the accuracy of the prediction.

[0035] Further, the common factor variance of each influence factor in the influence factor set is calculated.

[0036]

[0037] In the formula, r represents the common factor variance of the influence factor, X represents the input data of the influence factor, and Y represents the common factor data of the influence factor.

[0038] As can be known from the above description, the common factor variance reflects the influence degree of the influencing factors on the project cost, and can effectively and accurately screen the features.

[0039] Further, the PLS-ANN integrated prediction model based on the characteristic quantity is used to predict the cost of the electric-hydrogen coupling conversion grid sending project according to different voltage levels, and a prediction result is obtained.

[0040] For the electric-hydrogen coupling conversion grid sending project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the PLS-ANN integrated prediction model is used to predict the cost based on the characteristic quantity, and a first prediction result is obtained.

[0041] For the electric-hydrogen coupling conversion grid sending project with a voltage level of 35kV-220kV, the back propagation algorithm in the PLS-ANN integrated prediction model is used to predict the cost based on the characteristic quantity, and a second prediction result is obtained.

[0042] As can be known from the above description, since the project cost of the electric-hydrogen coupling conversion grid sending project is closely related to the voltage level, the higher the voltage level, the higher the required insulation level, the more expensive the equipment, the more funds required, and the cost level will be increased. For the electric-hydrogen coupling conversion grid sending project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the model can be used to predict the cost, and for the electric-hydrogen coupling conversion grid sending project with a voltage level of 35kV-220kV, the back propagation algorithm in the model can be used to predict the cost. Through differential prediction analysis, the project cost of the electric-hydrogen coupling conversion grid sending project can be more reasonably, comprehensively and accurately predicted.

[0043] Please refer to Figure 2 An electric-hydrogen coupling conversion grid sending project cost prediction terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0044] determining an influencing factor set of the project cost of the electric-hydrogen coupling conversion grid sending project;

[0045] extracting features from the influencing factor set using factor analysis to obtain characteristic quantities;

[0046] using a PLS-ANN integrated prediction model based on the characteristic quantities to predict the cost of the electric-hydrogen coupling conversion grid sending project according to different voltage levels, and obtaining a prediction result.

[0047] From the above description, the beneficial effects of the present application are that the influence factor set of the construction cost of the electric-hydrogen coupling conversion power grid sending-out project is determined, the influence factor set is characterized by using the factor analysis method, the characteristic quantity is obtained, the PLS-ANN integrated prediction model is used based on the characteristic quantity to predict the construction cost of the electric-hydrogen coupling conversion power grid sending-out project according to the different voltage grades, the most relevant influence factor of the construction cost is effectively determined by using the factor analysis method, the influence factor is used as the characteristic quantity, the PLS-ANN integrated prediction model is used to predict the construction cost according to the different voltage grades, the differentiated prediction is realized, different algorithms are combined, the advantages of different algorithms are combined, and therefore more accurate and effective construction cost prediction is realized.

[0048] Further, the electric-hydrogen coupling conversion power grid sending-out project includes an electric-hydrogen coupling conversion power grid sending-out power generation project and an electric-hydrogen coupling conversion power grid sending-out sending-out project.

[0049] The influence factor set of the construction cost of the electric-hydrogen coupling conversion power grid sending-out project includes:

[0050] The total station building area, the total station land area, the main control building area, the land area in the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the circuit breaker unit price, the main transformer unit price, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line, and the ground resistance are determined as the first influence factor set of the electric-hydrogen coupling conversion power grid sending-out power generation project.

[0051] The line folding single length, the tower material amount, the complex terrain length, the total tower base, the wire material amount, the double circuit length, the strain angle tower base, the total earthwork amount, the foundation steel material amount, the single wire area, the single circuit length, the wire price, and the foundation steel price are determined as the second influence factor set of the electric-hydrogen coupling conversion power grid sending-out sending-out project.

[0052] As can be seen from the above description, according to the engineering cost experience, the hydrogen coupling conversion power grid sending out power generation project is greatly affected by the station site geological conditions and the main equipment model, and is relatively less affected by the meteorological conditions, so the total station building area, the total station land area, the main control building land area, the land area within the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the single price of the circuit breaker, the single price of the main transformer, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line and the ground resistance are determined as the first influence factor set of the hydrogen coupling conversion power grid sending out power generation project, and the hydrogen coupling conversion power grid sending out project is affected by the path factor, the meteorological condition, the conductor model, the tower shape and the tower height four technical and other technical conditions, the line folding length, the tower material amount, the complex terrain length, the total number of poles, the wire material amount, the double circuit length, the strain angle tower base number, the total amount of earthwork, the foundation steel material amount, the single conductor area, the single circuit length, the conductor price and the foundation steel price are determined as the second influence factor set of the hydrogen coupling conversion power grid sending out project, and the preliminary confirmation of the engineering cost influence factors is realized.

[0053] Further, the influence factor set is extracted by using the factor analysis method, and the characteristic quantity includes:

[0054] The common factor variance of each influence factor in the influence factor set is calculated.

[0055] The influence factor set is sorted according to the order from large to small of the common factor variance, and the sorted influence factor set is obtained.

[0056] A preset number of influence factors are selected from the sorted influence factor set as characteristic quantities.

[0057] As can be seen from the above description, the hydrogen coupling conversion power grid sending out project cost data sample contains many influence factors, and the input data of the prediction model has the problems of high data dimension and difficult fitting, which cannot realize effective prediction of the engineering cost, so the factor analysis method is used to determine the characteristic quantity by calculating the common factor variance of each influence factor, the characteristic quantity is the characteristic data which greatly affects the prediction accuracy of the model, so as to reduce the data dimension, thereby ensuring the accuracy of the prediction.

[0058] Further, the common factor variance of each influence factor in the influence factor set is calculated.

[0059]

[0060] In the formula, r represents the common factor variance of the influence factor, X represents the input data of the influence factor, and Y represents the common factor data of the influence factor.

[0061] From the above description, it can be seen that the common factor variance reflects the influence degree of the influencing factors on the project cost, and can effectively and accurately screen the characteristics.

[0062] Further, the cost of the electric-hydrogen coupling conversion grid sending project is predicted according to different voltage levels using the PLS-ANN integrated prediction model based on the characteristic quantity, and the prediction result includes:

[0063] For the electric-hydrogen coupling conversion grid sending project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the PLS-ANN integrated prediction model is used to predict the cost based on the characteristic quantity, and a first prediction result is obtained.

[0064] For the electric-hydrogen coupling conversion grid sending project with a voltage level of 35kV-220kV, the back propagation algorithm in the PLS-ANN integrated prediction model is used to predict the cost based on the characteristic quantity, and a second prediction result is obtained.

[0065] From the above description, it can be seen that the cost of the electric-hydrogen coupling conversion grid sending project is closely related to the voltage level. The higher the voltage level, the higher the required insulation level, the more expensive the equipment, the more investment required, and the higher the cost level. For the electric-hydrogen coupling conversion grid sending project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the model can be used to predict the cost. For the electric-hydrogen coupling conversion grid sending project with a voltage level of 35kV-220kV, the back propagation algorithm in the model can be used to predict the cost. Through differential prediction analysis, the cost of the electric-hydrogen coupling conversion grid sending project can be more reasonably, comprehensively and accurately predicted.

[0066] The above-mentioned electric-hydrogen coupling conversion grid sending project cost prediction method and terminal of the application can be applied to the electric-hydrogen coupling conversion grid sending project cost prediction scene. The following will be described through specific embodiments:

[0067] Please refer to Figure 1 and Figure 3 , the embodiment one of the application is:

[0068] An electric-hydrogen coupling conversion grid sending project cost prediction method, comprising the steps of:

[0069] S1, determining the influencing factor set of the electric-hydrogen coupling conversion grid sending project cost, specifically including S11-S12:

[0070] The electric-hydrogen coupling conversion grid sending project includes an electric-hydrogen coupling conversion grid sending power generation project and an electric-hydrogen coupling conversion grid sending sending project.

[0071] The cost level of the electric hydrogen coupling conversion power grid sending out project is related to the price of construction materials required in the process of project construction, the amount of project completed, labor, construction measures and the amount of fee paid. The formation of the project cost is influenced by a number of internal and external factors, and the formation mechanism is relatively complex. Therefore, in order to predict the project cost, the main factors influencing the project cost need to be determined. Since the payment has great particularity, the present application mainly starts from the factors of the amount of project and the price of materials to analyze the main construction conditions influencing the cost of the electric hydrogen coupling conversion power grid sending out project.

[0072] S11, the total station building area, the total station land area, the main control building area, the land area inside the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the unit price of the circuit breaker, the unit price of the main transformer, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line and the ground resistance are determined as the first influence factor set of the electric hydrogen coupling conversion power grid sending out power generation project.

[0073] The first influence factor set is determined because according to the experience of the project cost, the electric hydrogen coupling conversion power grid sending out power generation project is greatly influenced by the station site geological conditions and the main equipment type, and is relatively less influenced by the meteorological conditions.

[0074] S12, the line folding length, the tower material amount, the complex terrain length, the total tower base number, the wire material amount, the double circuit length, the strain angle tower base number, the total earthwork amount, the foundation steel material amount, the single wire area, the single circuit length, the wire price and the foundation steel price are determined as the second influence factor set of the electric hydrogen coupling conversion power grid sending out project. The complex terrain includes river network, marsh, mountain, high mountain, desert and steep ridge.

[0075] Unlike the electric hydrogen coupling conversion power grid sending out power generation project, the electric hydrogen coupling conversion power grid sending out project is mainly influenced by the path factor, the meteorological condition, the wire type, the tower shape and the tower height, so the above factors are taken as the second influence factor.

[0076] The cost data sample of the electric hydrogen coupling conversion power grid sending out project contains many influencing factors, and as the input data of the prediction model, the data dimension is high and fitting is difficult. That is, when the data dimension is too large, the data attribute is greater than the data sample number, even if the model fitting method for small sample prediction problem is not accurate, which leads to over data fitting, reduces the generalization ability of the prediction model, at the same time, the high dimension characteristics of the data increase the complexity of the calculation, which is not convenient for actual engineering analysis and prediction fitting analysis model. Therefore, the feature extraction is carried out by the factor analysis method, the feature data which greatly influences the prediction accuracy of the model is extracted, and the data dimension is reduced, which is as follows:

[0077] S2, performing feature extraction on the set of influence factors using a factor analysis method to obtain characteristic quantities, specifically comprising S21-S23:

[0078] S21, calculating a common factor variance of each influence factor in the set of influence factors, specifically comprising:

[0079]

[0080] wherein r represents the common factor variance of the influence factor, X represents the input data of the influence factor, and Y represents the common factor data of the influence factor, which is obtained by performing a factor calculation on the original data of the influence factor.

[0081] S22, sorting the set of influence factors in descending order of the common factor variance to obtain a sorted set of influence factors.

[0082] S23, selecting a preset number of influence factors from the sorted set of influence factors as characteristic quantities.

[0083] The preset number is set according to actual needs.

[0084] For example, taking 220kV as an example for the electric-hydrogen coupling conversion power grid sending-out power generation project, the factor analysis result is shown in Table 1, the result shows that the total station building area has the highest common factor variance, followed by the total station land area, assuming that the preset number is 8, the total station land area, the total station land area, the main control building land area, the land area inside the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the single price of the circuit breaker and the single price of the main transformer are selected as characteristic quantities.

[0085] Table 1 Common factor variance of the first influence factor of the electric-hydrogen coupling conversion power grid sending-out power generation project

[0086]

[0087]

[0088] Taking 500kV as an example for the electric-hydrogen coupling conversion power grid sending-out sending-out project, the factor analysis result is shown in Table 2, the result shows that the line single length has the highest common factor variance, followed by the tower material quantity, assuming that the preset number is 8, the line single length, the tower material quantity, the complex terrain length, the total tower base, the wire quantity, the double circuit length, the strain angle tower base and the total earthwork quantity are selected as characteristic quantities.

[0089] Table 2 Common factor variance of the second influence factor of the electric-hydrogen coupling conversion power grid sending-out sending-out project

[0090] Characteristic quantity Common factor variance Line folding length 0.9836 Tower material amount 0.9798 Complex terrain length 0.9759 Total tower base 0.9751 Wire material amount 0.9713 Double circuit length 0.9523 Tension angle tower base 0.9492 Earthwork total amount 0.8866 Foundation steel material amount 0.7809 Single wire area 0.7608 Single circuit length 0.7570 Wire price 0.6712 Foundation steel-price 0.6335

[0091] According to previous experience, the engineering cost of the electric-hydrogen coupling conversion power grid sending project is closely related to the voltage level. The higher the voltage level, the higher the required insulation level, the more expensive the equipment, the more investment required, and the higher the cost level. Therefore, the cost prediction problem should be analyzed according to the different characteristics of the prediction model according to the voltage level, and different models are used for prediction. In view of this characteristic of the electric-hydrogen coupling conversion power grid sending project cost prediction problem, the idea of ensemble learning (EL) is used, which is to divide the analyzed data samples according to different characteristics, apply different statistical learning methods to analyze these data according to the characteristics of different samples, and finally integrate the information obtained by different algorithms, as follows:

[0092] S3, using the PLS-ANN integrated prediction model based on the characteristic quantity to predict the cost of the electric-hydrogen coupling conversion power grid sending project according to the different voltage levels, to obtain the prediction result, as shown in Figure 3 , specifically including S31-S32:

[0093] S31, for the electric-hydrogen coupling conversion power grid sending project with voltage level of 330kV-1000kV, using the partial least squares regression (PLSR) algorithm in the PLS-ANN integrated prediction model based on the characteristic quantity to predict the cost, to obtain the first prediction result.

[0094] Wherein, the PLSR algorithm process is:

[0095] (1) Determine the objective function:

[0096]

[0097] In the formula, represents the rank of the column matrix, B0 represents the matrix of the row matrix, represents the rank of the row matrix, A0 represents the column matrix, W1 represents the unit eigenvector corresponding to the maximum eigenvalue of the first matrix, represents the partial least squares regression parameter, T1 represents the unit eigenvector corresponding to the maximum eigenvalue of the second matrix. In this embodiment, T1 is a constant, ||T1||=1, so T1=1, G1=B0, which can be solved:

[0098]

[0099] In the formula, G1 represents the transition matrix calculated.

[0100] Since W1 and T1 are unit vectors, we have:

[0101]

[0102]

[0103]

[0104] [Cov(x1,y)A 01 +Cov(x2,y)A 02 +...+Cov(x k ,y)A 0k ];

[0105] In the formula, A 0k Let y represent the dependent variable and x1 represent the independent variable 1. p Let p and x represent the independent variables. j Let j and x represent independent variables. k Let k and x represent the independent variables. i Let i represent the independent variable.

[0106] The regression relationship between A0 and B0 on ​​F1 is as follows:

[0107]

[0108] In the formula, Let r1 denote the rank transformation of the solution vector, r1 denote the solution vector, A1 denotes column matrix 1, B1 denotes row matrix 1, and F1 denotes principal component 1.

[0109] (2) Cross-validation

[0110] The formula for principal component crossover test is:

[0111]

[0112] In the formula, S i Representing residuals i and y h Represents the dependent variable h. PRESS represents the mirror image of the dependent variable. i Represents the step residual, y i Indicates the dependent variable i, Represents the mirror image of the dependent variable h, S i-1 Represents residual i-1, represents the standard deviation, and n represents the ordinal number of the variable.

[0113] when If introducing a new principal component F2 can improve the fit of the regression equation, then A1 is used to replace A0 to recalculate the principal component F2.

[0114] (3) Establishing the regression equation

[0115] Once the principal components meet the accuracy requirement, the least squares linear regression equation is used to fit the equation, resulting in:

[0116] B0 = r1F1 + r2F2 + ... + r h F h ;

[0117] In the formula, r h It means that F h express.

[0118] Due to F i All are linear combinations of standardized variables, and according to the PLS component definition, F i =A i-1 W i Introduce variable W i * :

[0119]

[0120] In the formula, W k express, express.

[0121] The above equation can be simplified to:

[0122] F i =A0W i * ;

[0123] Let b i =W i * r Substitute In the middle, the regression equation is simplified and reduced to standardized variables:

[0124]

[0125] Further, let's write the above equation as an equation with the original variables:

[0126]

[0127] In the formula, Let E(y) represent the dependent variable, and b represent the expected value of the dependent variable. i Represents the regression coefficients i, s y Let y represent the regression variance. Let i represent the regression variance, E(x) i ) represents the expected value of independent variable i, b m Represents the regression coefficients m, x m Let m represent the independent variable. This equation is used to predict construction costs.

[0128] S32, for the voltage level of 35kV-220kV electric hydrogen coupling conversion grid sending project, based on the characteristic quantity, using the back propagation (BP) algorithm (an artificial neural network algorithm) in the PLS-ANN integrated prediction model to predict the cost, and a second prediction result is obtained.

[0129] Wherein, the BP network structure is: input layer d neurons, hidden layer n H neurons, output layer c neurons, network weight w.

[0130] Input training sample x=(x1,…,x d ) T , expected output t=(t1,…,t c ) T , network actual output z=(z1,…,z c ) T , hidden layer neuron output net is the net output of neurons, and has:

[0131] Hidden layer: y j =f(net j ), i=1,…,n H ;

[0132] Output layer: z k =f(net k ), k=1,…,c;

[0133] The error sum of squares criterion function is J(w):

[0134]

[0135] Wherein, t k represents the base of measurement, z k represents the measurement control, and c represents the ordinal number.

[0136] The above engineering cost prediction method of the application is analyzed by examples, first, the characteristic quantity is input, including the characteristic index value of each type of voltage level, and the cost is predicted by the PLS-ANN integrated prediction model, and the prediction result shown in table 3 is obtained.

[0137] Table 3 prediction analysis result

[0138]

[0139] It can be seen that the deviation between the data predicted by the electric hydrogen coupling conversion grid sending project cost prediction method of the application and the true data is small, and more accurate and effective engineering cost prediction is realized.

[0140] Please refer to Figure 2 Embodiment two of the present invention is as follows:

[0141] A terminal for predicting the cost of an electro-hydrogen coupling conversion power grid transmission project includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the electro-hydrogen coupling conversion power grid transmission project cost prediction method in Embodiment 1.

[0142] In summary, this invention provides a method and terminal for predicting the cost of an electric-hydrogen coupling conversion power grid transmission project. It identifies the set of influencing factors for the project cost, uses factor analysis to extract features from this set, and then uses a PLS-ANN ensemble prediction model to predict the cost of the project according to different voltage levels. This method effectively identifies the most relevant influencing factors for project cost using factor analysis, uses these factors as features, and employs the PLS-ANN ensemble prediction model to predict project cost according to different voltage levels. This achieves differentiated prediction and integrates the advantages of different algorithms, resulting in more accurate and effective cost prediction. Furthermore, by using factor analysis to calculate the common factor variance of each influencing factor to determine the features, these features significantly improve the accuracy of the prediction model, thus reducing data dimensionality and ensuring prediction accuracy.

[0143] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for predicting the cost of an electrical hydrogen coupling conversion grid transmission project, characterized in that, The method comprises the steps of: determining a set of influence factors of project cost of the electric-hydrogen coupling conversion power grid sending-out project; extracting features of the set of influence factors by using a factor analysis method to obtain characteristic quantities; predicting the project cost of the electric-hydrogen coupling conversion power grid sending-out project according to different voltage levels by using a PLS-ANN integrated prediction model based on the characteristic quantities to obtain a prediction result; the electric-hydrogen coupling conversion power grid sending-out project comprises an electric-hydrogen coupling conversion power grid sending-out power generation project and an electric-hydrogen coupling conversion power grid sending-out project; the set of influence factors of project cost of the electric-hydrogen coupling conversion power grid sending-out project comprises: the total station building area, the total station land area, the main control building area, the land area within the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the circuit breaker unit price, the main transformer unit price, the single main transformer capacity, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line, and the ground resistance are determined as the first set of influence factors of the electric-hydrogen coupling conversion power grid sending-out power generation project; the line folding single length, the tower material amount, the complex terrain length, the total tower base, the wire material amount, the double circuit length, the strain angle tower base, the total earthwork amount, the foundation steel material amount, the single wire area, the single circuit length, the wire price, and the foundation steel price are determined as the second set of influence factors of the electric-hydrogen coupling conversion power grid sending-out project; the step of extracting features of the set of influence factors by using a factor analysis method to obtain characteristic quantities comprises: calculating the common factor variance of each influence factor in the set of influence factors; sorting the set of influence factors in descending order of the common factor variance to obtain a sorted set of influence factors; selecting a preset number of influence factors from the sorted set of influence factors as characteristic quantities; the step of calculating the common factor variance of each influence factor in the set of influence factors comprises: ； wherein r represents the common factor variance of the influence factor, X represents the input data of the influence factor, and Y represents the common factor data of the influence factor; the step of predicting the project cost of the electric-hydrogen coupling conversion power grid sending-out project according to different voltage levels by using a PLS-ANN integrated prediction model based on the characteristic quantities to obtain a prediction result comprises: for the electric-hydrogen coupling conversion power grid sending-out project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the PLS-ANN integrated prediction model is used to predict the project cost based on the characteristic quantities to obtain a first prediction result; for the electric-hydrogen coupling conversion power grid sending-out project with a voltage level of 35kV-220kV, the back propagation algorithm in the PLS-ANN integrated prediction model is used to predict the project cost based on the characteristic quantities to obtain a second prediction result.

2. An electric hydrogen coupling conversion grid sending project cost prediction terminal, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the following steps: determining a set of influence factors of project cost of the electric-hydrogen coupling conversion power grid sending-out project; extracting features of the set of influence factors by using a factor analysis method to obtain characteristic quantities; predicting the project cost of the electric-hydrogen coupling conversion power grid sending-out project according to different voltage levels by using a PLS-ANN integrated prediction model based on the characteristic quantities to obtain a prediction result; The electric-hydrogen coupling conversion power grid sending project comprises an electric-hydrogen coupling conversion power grid sending power generation project and an electric-hydrogen coupling conversion power grid sending project; The set of influence factors for determining the engineering cost of the electric-hydrogen coupling conversion power grid sending project comprises: The total building area, the total land area, the main control building area, the land area within the fence, the number of circuit breakers, the high-voltage side outgoing line scale, the unit price of the circuit breaker, the unit price of the main transformer, the capacity of a single main transformer, the low-voltage side outgoing line, the station site excavation amount, the medium-voltage side outgoing line, and the ground resistance are determined as the first set of influence factors of the electric-hydrogen coupling conversion power grid sending power generation project; The line folding length, the tower material amount, the complex terrain length, the total number of tower bases, the wire material amount, the double-circuit length, the strain angle tower base number, the total amount of earthwork, the foundation steel material amount, the single wire area, the single-circuit length, the wire price, and the foundation steel price are determined as the second set of influence factors of the electric-hydrogen coupling conversion power grid sending project; The characteristic quantity obtained by using the factor analysis method to perform feature extraction on the set of influence factors comprises: Calculating the common factor variance of each influence factor in the set of influence factors; sequentially sorting the set of influence factors in descending order of the common factor variances to obtain a sorted set of influence factors; selecting a preset number of influence factors from the sorted set of influence factors as characteristic quantities; The calculation of the common factor variance of each influence factor in the set of influence factors comprises: ； wherein r represents the common factor variance of the influence factor, X represents the input data of the influence factor, and Y represents the common factor data of the influence factor; The cost prediction of the electric-hydrogen coupling conversion power grid sending project according to different voltage levels based on the characteristic quantities using the PLS-ANN integrated prediction model comprises: For the electric-hydrogen coupling conversion power grid sending project with a voltage level of 330kV-1000kV, the partial least squares regression algorithm in the PLS-ANN integrated prediction model is used to perform cost prediction based on the characteristic quantities, and a first prediction result is obtained; For the electric-hydrogen coupling conversion power grid sending project with a voltage level of 35kV-220kV, the back propagation algorithm in the PLS-ANN integrated prediction model is used to perform cost prediction based on the characteristic quantities, and a second prediction result is obtained.

Images