Urban energy consumption level and urban ecology-development coupling degree calculation method based on noctilucent remote sensing

Abstract

The invention discloses an urban energy consumption level and urban ecology-development coupling degree calculation method based on noctilucent remote sensing, and relates to the field of remote sensing and urban development, and the method comprises the steps: obtaining the gridded noctilucent remote sensing data, urban ecology data and urban development data of a target area; based on the noctilucent remote sensing data, utilizing a pre-trained neural network model to identify urban energy consumption data of each grid; calculating an energy consumption index according to the urban energy consumption data, and respectively calculating an ecological index and a development index based on the weight of each data in the urban ecological data and the urban development data; and according to the energy consumption index, the ecological index and the development index, calculating the urban energy consumption level and the urban ecology-development coupling degree. According to the invention, the refinement level of urban energy consumption monitoring is improved, and an innovative technical means is provided for urban complex system research and sustainable development management.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of remote sensing and urban development, and more particularly to a method for calculating urban energy consumption level and urban ecological-development coupling degree based on night light remote sensing. BACKGROUND

[0002] The energy consumption level of a city reflects the development of the city to some extent, and the environmental pollution caused by energy consumption affects the regional ecological system, thereby restricting sustainable development. Therefore, it is of great significance to explore the coordination relationship among energy, ecology and development.

[0003] However, in previous related research, energy consumption mainly depends on the statistics of national and local departments, which has great limitations in research scale, and there are still a large number of data missing in some areas, making it difficult to monitor the energy consumption level in detail and restricting the in-depth exploration of the interaction between energy consumption and urban ecological-development system. Moreover, most of the existing researches focus on the relationship between urban energy consumption and urban ecology, and the relationship between urban ecology and urban development, lacking research on the complex coupling relationship among urban energy consumption, ecology and development.

[0004] Therefore, how to break through the limitations of traditional statistical data and analyze and calculate the complex coupling relationship among urban energy consumption, ecology and development is a problem that needs to be solved by those skilled in the art. SUMMARY

[0005] Therefore, the present application provides a method for calculating urban energy consumption level and urban ecological-development coupling degree based on night light remote sensing to solve the above problems.

[0006] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0007] The present application discloses a method for calculating urban energy consumption level and urban ecological-development coupling degree based on night light remote sensing, comprising:

[0008] Obtaining grid night light remote sensing data, urban ecological data and urban development data of a target area;

[0009] Based on the night light remote sensing data, using a pre-trained neural network model to identify the urban energy consumption data of each grid;

[0010] According to the urban energy consumption data, calculating an energy consumption index, and based on the weights of each data in the urban ecological data and the urban development data, respectively calculating an ecological index and a development index;

[0011] According to the energy consumption index, the ecological index and the development index, calculating the urban energy consumption level and the urban ecological-development coupling degree.

[0012] Further, the urban ecological data includes: annual average concentration of sulfur dioxide, annual average concentration of nitrogen dioxide, industrial solid waste discharge, industrial wastewater discharge, water resources per capita, green coverage rate, wetland area;

[0013] The urban development data includes: population natural growth rate, population density, per capita GDP, employment rate, total industrial output value, and proportion of tertiary industry.

[0014] Further, the neural network model is a back propagation neural network, and the back propagation neural network includes an input layer, a hidden layer, and an output layer.

[0015] The number of neurons in the input layer is the same as the dimension of the input feature vector, and is used for inputting sample features.

[0016] The number of layers and the number of neurons in each layer of the hidden layer are determined by a genetic algorithm.

[0017] The output layer is a single neuron, and is used for outputting the urban energy consumption prediction value of the corresponding grid.

[0018] Further, the training set and the test set of the back propagation neural network are obtained in the following manner:

[0019] From the noctilucent remote sensing data, a plurality of grids are randomly selected, and noctilucent remote sensing features and annual total energy consumption in each grid are counted; the noctilucent remote sensing features include average brightness value, median brightness value, brightness standard deviation, and brightness discrete value in the grid.

[0020] A feature vector formed by the noctilucent remote sensing features is used as sample input features, and the annual total energy consumption is used as sample output features to form a training sample. A plurality of training samples are constructed based on the plurality of selected grids, and the training set and the test set are divided according to an 8:2 ratio. The back propagation neural network is trained and tested.

[0021] Further, the formula of the brightness discrete value is:

[0022]

[0023] w ij = 1 / d ij ;

[0024] wherein DiscreteI represents the brightness discrete value of the grid, N represents the total number of pixels in the grid, I i , I j respectively represent the brightness value of pixel i and pixel j in the grid, represents the average brightness value of the pixels in the grid, w ij represents the weight of pixel i and pixel j, and dij This represents the pixel distance between pixel i and pixel j.

[0025] Furthermore, the number of hidden layers and the number of neurons in each layer are determined through optimization using a genetic algorithm, specifically including:

[0026] Step 1: Set the initial population size and the maximum number of iterations;

[0027] Step 2: Perform population initialization, where the individual gene sequence corresponds one-to-one with the number of neurons in each hidden layer;

[0028] Step 3: Calculate the fitness value of individuals in the population and identify the optimal individual. Determine whether the preset stopping condition is met or the maximum number of iterations has been reached. If the stopping condition is met or the maximum number of iterations has been reached, stop the iteration; otherwise, proceed to step 4.

[0029] Step 4: Remove individuals with fitness values ​​lower than the average in the population, and calculate the mutation probability and crossover probability for each individual;

[0030] Step 5: Perform mutation operations on the remaining individuals in the population according to the mutation probability to generate new individuals;

[0031] Step 6: Perform crossover operation on the remaining individuals in the population according to the crossover probability to generate new individuals. Add the new individuals generated in Step 5 and Step 6 to the population and return to Step 3.

[0032] The preset stopping condition is that the fitness value of the optimal individual is greater than a preset value, or the variance of the population fitness value is less than a preset value; the fitness value is the reciprocal of the prediction error of the backpropagation neural network trained after determining the number and quantity of hidden layer neurons based on the individual gene sequence.

[0033] Furthermore, the weights of each data point in the urban ecological data and urban development data are calculated based on subjective and objective weights, using the following formula:

[0034]

[0035] Among them, w i w represents the weight of the i-th data item. AHP,i w e,i These are the subjective and objective weights of the i-th data item, respectively; w AHP,j w e,j α and β represent the subjective and objective weights of the j-th data item in the k-th data items of the same type, respectively; α and β are both adjustment coefficients.

[0036] Furthermore, the subjective weights are determined using the analytic hierarchy process (AHP), specifically including:

[0037] A hierarchical model is established, which includes a target layer, a criterion layer, and an indicator layer;

[0038] Based on the established hierarchical model, a judgment matrix is ​​constructed by assigning values ​​using the scaling method by domain experts;

[0039] Perform a consistency check on the indicator weights of the constructed judgment matrix;

[0040] After the consistency test is passed, the weights of each data item in the indicator layer are calculated based on the weights between the target layer and the criterion layer, as well as the weights between the criterion layer and the indicator layer.

[0041] Furthermore, the objective weights are calculated based on the information entropy of each data point, specifically including:

[0042] A sample dataset is constructed using each data point in any grid within the urban ecological data / urban development data as a sample.

[0043] Standardize the data of each sample in the sample dataset;

[0044] Calculate the information entropy of each data point;

[0045] The corresponding weights are determined based on the information entropy of each data point.

[0046] Furthermore, the calculation formulas for the energy consumption index, the ecological index, and the development index are as follows:

[0047]

[0048] Among them, U a This represents energy consumption indicators, where a1 and a2 represent total energy consumption and per capita energy consumption, respectively. These are the corresponding weights; U b Indicating ecological indicators, b i This represents the i-th data point in n urban ecological data points. For the corresponding weights; U c c represents a development indicator. i This represents the i-th data item in m city development data items. The corresponding weights;

[0049] The formula for calculating the urban energy consumption level and the urban ecology-development coupling degree C is as follows:

[0050]

[0051] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing, which has the following technical effects:

[0052] This invention, based on nighttime light remote sensing data, uses a pre-trained backpropagation neural network to invert energy consumption data for each grid. Leveraging the spatial continuity of nighttime light remote sensing, it improves energy consumption monitoring accuracy to the grid scale, covering regions at different administrative levels and effectively filling monitoring gaps in areas with missing data, providing data support for refined energy consumption management. Furthermore, the neural network model optimizes the hidden layer structure through a genetic algorithm and is trained using nighttime light remote sensing features and measured energy consumption data, accurately capturing the correlation between nighttime light radiation and energy consumption. The introduction of discrete brightness values ​​further considers pixel spatial relationships, enhancing the model's ability to invert energy consumption on complex underlying surfaces, resulting in significantly better spatial resolution and accuracy of energy consumption data compared to traditional statistical methods. Simultaneously, this invention calculates the weights of ecological and development indicators by combining subjective and objective weights, incorporating expert experience while respecting data information, avoiding the bias of single-weighting methods. Based on this, it achieves a quantitative expression of the complex relationship among these three factors, providing an operable quantitative tool for analyzing the synergistic mechanism of energy consumption, ecological environment, and urban development. This invention improves the precision of urban energy consumption monitoring, establishes a quantitative analysis framework for the coupling relationship between energy consumption, ecology and development, and provides innovative technical means for the study of complex urban systems and sustainable development management. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] This invention discloses a method for calculating the coupling degree between urban energy consumption level and urban ecology-development based on nighttime light remote sensing, such as... Figure 1 As shown, it includes:

[0057] Acquire gridded nighttime light remote sensing data, urban ecological data, and urban development data for the target area;

[0058] Based on nighttime light remote sensing data, a pre-trained neural network model is used to identify urban energy consumption data for each grid.

[0059] Energy consumption indicators are calculated based on urban energy consumption data, and ecological and development indicators are calculated based on the weights of various data in urban ecological and urban development data.

[0060] Based on energy consumption indicators, ecological indicators, and development indicators, the coupling degree between urban energy consumption level and urban ecology-development is calculated.

[0061] In one specific embodiment, urban ecological data includes: annual average concentration of sulfur dioxide, annual average concentration of nitrogen dioxide, industrial solid waste discharge, industrial wastewater discharge, per capita water resources, green coverage rate, and wetland area.

[0062] Urban development data includes: natural population growth rate, population density, GDP per capita, employment rate, total industrial output, and the proportion of the tertiary industry.

[0063] Specifically, by rationally setting the grid size and based on existing urban ecological and development statistics, the annualized average values ​​of urban ecological and urban development data within the grid are calculated. The nighttime light remote sensing data consists of multiple nighttime light remote sensing images corresponding to the urban ecological and development data for the corresponding year, which are then aligned pixel-by-pixel, and the average brightness values ​​of each pixel are calculated to obtain the final nighttime light remote sensing image.

[0064] In one specific embodiment, the neural network model is a backpropagation neural network, which includes an input layer, a hidden layer, and an output layer;

[0065] The number of neurons in the input layer is the same as the dimension of the input feature vector, and they are used to input sample features;

[0066] The number of hidden layers and the number of neurons in each layer are determined by optimization using a genetic algorithm;

[0067] The output layer consists of a single neuron, which outputs the predicted urban energy consumption value for the corresponding grid.

[0068] In one specific embodiment, the training and test sets of the backpropagation neural network are obtained in the following way:

[0069] From nighttime light remote sensing data, several grids were randomly selected, and the nighttime light remote sensing characteristics and annual energy consumption within each grid were statistically analyzed. The nighttime light remote sensing characteristics included the average brightness value, median brightness value, standard deviation of brightness value, and brightness dispersion value within the grid. Specifically, energy consumption data mainly relies on statistics from national and local departments, mostly representing statistics for a specific city or urban administrative region. This data is not comprehensive for more detailed regional statistics, and some regions have missing data, resulting in significant limitations in the research scale. Therefore, data statistics were performed on grid areas with complete and detailed energy consumption data to obtain nighttime light remote sensing characteristics and annual energy consumption, and per capita energy consumption was calculated based on the annual energy consumption.

[0070] The feature vectors formed by various nighttime light remote sensing features are used as the input features of the samples, and the total annual energy consumption is used as the output features of the samples to form a training sample. Several training samples are constructed based on several selected grids, and the training set and test set are obtained by dividing them in an 8:2 ratio. The backpropagation neural network is then trained and tested.

[0071] In one specific embodiment, the formula for the luminance discrete value is:

[0072]

[0073] w ij =1 / d ij ;

[0074] Where DiscreteI represents the discrete value of the grid's brightness, N represents the total number of pixels in the grid, and I i I j These represent the brightness values ​​of pixels i and j within the grid, respectively. w represents the average pixel brightness within the grid. ij d represents the weights of pixels i and j. ij This represents the pixel distance between pixel i and pixel j.

[0075] In a specific embodiment, the number of hidden layers and the number of neurons in each layer are determined through optimization using a genetic algorithm, specifically including:

[0076] Step 1: Set the initial population size and the maximum number of iterations;

[0077] Step 2: Perform population initialization, where the individual gene sequence corresponds one-to-one with the number of neurons in each hidden layer;

[0078] Step 3: Calculate the fitness value of individuals in the population and identify the optimal individual. Determine whether the preset stopping condition is met or the maximum number of iterations has been reached. If the stopping condition is met or the maximum number of iterations has been reached, stop the iteration; otherwise, proceed to step 4.

[0079] Step 4: Remove individuals with fitness values ​​lower than the average in the population, and calculate the mutation probability and crossover probability for each individual;

[0080] Step 5: Perform mutation operations on the remaining individuals in the population according to the mutation probability to generate new individuals; Step 6: Perform crossover operations on the remaining individuals in the population according to the crossover probability to generate new individuals, add the new individuals generated in Step 5 and Step 6 to the population, and return to Step 3.

[0081] Specifically, when performing mutation operations, firstly, the maximum and minimum mutation probabilities are determined based on the mutation probabilities of all individuals. Then, an individual is selected, and a random number between the minimum and maximum mutation probabilities is generated. If the mutation probability of the selected individual is less than the random number, the selected individual is not mutated; otherwise, the mutation operation is performed. This process is repeated for each selected individual, and each individual is checked for mutation to obtain several mutated individuals. Finally, a gene in the gene sequence of the mutated individual is randomly replaced with a random number in the gene's numerical space (for example, the second gene in the gene sequence corresponds to the number of neurons in the second layer of the hidden layer, and the numerical space of the second gene is [10, 30], which corresponds to the adjustable number range of neurons in the second layer). This process is repeated for each selected mutated individual to generate several new individuals.

[0082] During the crossover operation, firstly, the maximum and minimum mutation probabilities are determined based on the crossover probabilities of all individuals. Then, an individual is selected, and a random number between the minimum and maximum mutation probabilities is generated. If the mutation probability of the selected individual is less than the random number, the selected individual is not mutated; otherwise, the mutation operation is performed. This process is repeated for each selected individual, and it is determined whether each individual is mutated, resulting in several mutated individuals. Finally, gene length is determined, and individuals with the same gene length are randomly paired. One or two genes are randomly selected from each pair of individuals for gene crossover, resulting in two new individuals. This process is repeated for each selected crossover individual, generating several new individuals.

[0083] The formulas for calculating the mutation probability and crossover probability are as follows:

[0084]

[0085] Among them, f i P cro,i and P mut,i Let f represent the fitness value, crossover probability, and mutation probability of the i-th individual, respectively; avg f max f min Let P represent the average, maximum, and minimum fitness values ​​of the current population, respectively; e is the natural constant; P pop,cro P pop,mutThese represent the baseline crossover probability and mutation probability of the current population, respectively. g and g max s and s0 are the current iteration count and the maximum iteration count, respectively, and s and s0 are the current population size and the maximum population size, respectively.

[0086] The preset stopping condition is that the fitness value of the best individual is greater than the preset value, or the variance of the population fitness value is less than the preset value; the fitness value is the reciprocal of the prediction error of the backpropagation neural network trained after determining the number of hidden layer neurons based on the individual gene sequence.

[0087] In a specific embodiment, the weights of various data points in urban ecological data and urban development data are calculated based on subjective and objective weights, using the following formula:

[0088]

[0089] Among them, w i w represents the weight of the i-th data item. AHP,i w e,i These are the subjective and objective weights of the i-th data item, respectively; w AHP,j w e,j Let α and β represent the subjective and objective weights of the j-th data item in k similar data items (e.g., k is 7 when calculating the weights of data items in urban ecology data; k is 6 when calculating the weights of data items in urban development data), respectively; α and β are adjustment coefficients. The first part of the formula adjusts the influence of subjective and objective factors on the weights of each data item based on coefficient α; the second part of the formula, based on product normalization, further emphasizes data items with larger subjective and objective weights, thereby strengthening the weights of data items that are important to urban ecology or urban development.

[0090] In one specific embodiment, subjective weights are determined using the analytic hierarchy process (AHP), specifically including:

[0091] Establish a hierarchical model, which includes an objective layer, a criterion layer, and an indicator layer;

[0092] Based on the established hierarchical model, a judgment matrix is ​​constructed by assigning values ​​using the scaling method by domain experts;

[0093] Perform a consistency check on the indicator weights of the constructed judgment matrix;

[0094] After the consistency test is passed, the weights of each data item in the indicator layer are calculated based on the weights between the target layer and the criterion layer, as well as the weights between the criterion layer and the indicator layer.

[0095] Specifically, taking the analytic hierarchy process (AHP) to determine the weights of various data points in urban ecological data as an example, first, based on the annual average concentrations of sulfur dioxide and nitrogen dioxide, industrial solid waste emissions, industrial wastewater emissions, per capita water resources, green coverage rate, and wetland area, the data points for the indicator layer are determined. These data belong to environmental pollution and ecological resources, respectively; therefore, the criteria layer consists of environmental pollution criteria and ecological resource criteria, respectively. The target layer consists of ecological indicators. Then, experts in fields such as ecological environment and regional economy conduct pairwise comparisons of the relative importance of indicators at the same level, assigning values ​​using the 1-9 scale to construct a judgment matrix. For example, in the environmental pollution criteria, the importance of "industrial wastewater emissions" and "industrial solid waste emissions" is compared. If experts consider the former more important, a value of 6 can be assigned (indicating that the former is slightly more important than the latter). Next, the consistency index CI and the consistency ratio CR (CR = CI / RI, where RI is the average random consistency index) are calculated. If CR < 0.1, the judgment matrix passes the consistency test, and the weights are reasonable; otherwise, the judgment matrix needs to be adjusted until it meets the requirements. Finally, the weights are calculated layer by layer from the criterion layer to the indicator layer to obtain the total weight of each data point relative to the target layer. For example, the weight of "environmental pollution" in the target layer is determined first, and then the weights of various data points such as "industrial solid waste generation" under the criteria are determined.

[0096] In one specific embodiment, the objective weights are calculated based on the information entropy of each data point, specifically including:

[0097] A sample dataset is constructed using each data point in any grid within the urban ecological data / urban development data as a sample.

[0098] Standardize the data of each sample in the sample dataset;

[0099] Calculate the information entropy of each data point;

[0100] The corresponding weights are determined based on the information entropy of each data point.

[0101] Specifically, let the sample data x ir Let x represent the value of the i-th data item in the r-th sample. ir The data after standardization is y ir First, calculate the proportion of the i-th data item in the sample. Where l is the total number of samples, and ε is the minimum value, usually taken as 10. -6 To avoid y pq When the logarithm is 0, the logarithm is meaningless. Then, calculate the information entropy of the i-th data item. Finally, calculate the weight of the i-th data item.

[0102] In a specific embodiment, the calculation formulas for energy consumption indicators, ecological indicators, and development indicators are as follows:

[0103]

[0104] Among them, U a This represents energy consumption indicators, where a1 and a2 represent total energy consumption and per capita energy consumption, respectively. These are the corresponding weights; U b Indicating ecological indicators, b i This represents the i-th data point in n urban ecological data points. For the corresponding weights; U c c represents a development indicator. i This represents the i-th data item in m city development data items. The corresponding weights;

[0105] The formula for calculating the coupling degree C between urban energy consumption level and urban ecology-development is as follows:

[0106]

[0107] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0108] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating the coupling degree between urban energy consumption level and urban ecology-development based on nighttime light remote sensing, characterized in that, include: Acquire gridded nighttime light remote sensing data, urban ecological data, and urban development data for the target area; Based on the aforementioned nighttime light remote sensing data, a pre-trained neural network model is used to identify urban energy consumption data for each grid. Energy consumption indicators are calculated based on the urban energy consumption data, and ecological and development indicators are calculated based on the weights of various data in the urban ecological data and urban development data, respectively. The coupling degree between urban energy consumption level and urban ecology-development is calculated based on the energy consumption index, the ecological index, and the development index.

2. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 1, characterized in that, The urban ecological data includes: annual average concentration of sulfur dioxide, annual average concentration of nitrogen dioxide, industrial solid waste discharge, industrial wastewater discharge, per capita water resources, green coverage rate, and wetland area. The urban development data includes: natural population growth rate, population density, GDP per capita, employment rate, total industrial output, and the proportion of the tertiary industry.

3. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 1, characterized in that, The neural network model is a backpropagation neural network, which includes an input layer, a hidden layer, and an output layer; The number of neurons in the input layer is the same as the dimension of the input feature vector, and is used to input sample features; The number of hidden layers and the number of neurons in each layer are determined by optimization using a genetic algorithm; The output layer is a single neuron used to output the predicted urban energy consumption value for the corresponding grid.

4. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 3, characterized in that, The training and test sets of the backpropagation neural network are obtained in the following ways: From the aforementioned nighttime light remote sensing data, several grids are randomly selected, and the nighttime light remote sensing characteristics and annual energy consumption within each grid are statistically analyzed. The nighttime light remote sensing characteristics include the average brightness value, median brightness value, standard deviation of brightness value, and brightness dispersion value within the grid. A training sample is constructed using the feature vector formed by each nighttime light remote sensing feature and the total annual energy consumption as the sample output feature. Several training samples are constructed based on several selected grids and divided into training and testing sets in an 8:2 ratio. The backpropagation neural network is then trained and tested.

5. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 4, characterized in that, The formula for the discrete value of brightness is: w ij =1 / d ij ; Where DiscreteI represents the discrete value of the grid's brightness, N represents the total number of pixels in the grid, and I i I j These represent the brightness values ​​of pixels i and j within the grid, respectively. w represents the average pixel brightness within the grid. ij d represents the weights of pixels i and j. ij This represents the pixel distance between pixel i and pixel j.

6. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 3, characterized in that, The number of hidden layers and the number of neurons in each layer are determined through optimization using a genetic algorithm, specifically including: Step 1: Set the initial population size and the maximum number of iterations; Step 2: Perform population initialization, where the individual gene sequence corresponds one-to-one with the number of neurons in each hidden layer; Step 3: Calculate the fitness value of individuals in the population and identify the optimal individual. Determine whether the preset stopping condition is met or the maximum number of iterations has been reached. If the stopping condition is met or the maximum number of iterations has been reached, stop the iteration; otherwise, proceed to step 4. Step 4: Remove individuals with fitness values ​​lower than the average in the population, and calculate the mutation probability and crossover probability for each individual; Step 5: Perform mutation operations on the remaining individuals in the population according to the mutation probability to generate new individuals; Step 6: Perform crossover operation on the remaining individuals in the population according to the crossover probability to generate new individuals. Add the new individuals generated in Step 5 and Step 6 to the population and return to Step 3. The preset stopping condition is that the fitness value of the optimal individual is greater than a preset value, or the variance of the population fitness value is less than a preset value; the fitness value is the reciprocal of the prediction error of the backpropagation neural network trained after determining the number and quantity of hidden layer neurons based on the individual gene sequence.

7. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 1, characterized in that, The weights of each data point in the urban ecological data and urban development data are calculated based on subjective and objective weights, using the following formula: Among them, w i w represents the weight of the i-th data item. AHP,i w e,i These are the subjective and objective weights of the i-th data item, respectively; w AHP,j w e,j α and β represent the subjective and objective weights of the j-th data item in the k-th data items of the same type, respectively; α and β are both adjustment coefficients.

8. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 7, characterized in that, The subjective weights are determined using the analytic hierarchy process (AHP), specifically including: A hierarchical model is established, which includes a target layer, a criterion layer, and an indicator layer; Based on the established hierarchical model, a judgment matrix is ​​constructed by assigning values ​​using the scaling method by domain experts; Perform a consistency check on the indicator weights of the constructed judgment matrix; After the consistency test is passed, the weights of each data item in the indicator layer are calculated based on the weights between the target layer and the criterion layer, as well as the weights between the criterion layer and the indicator layer.

9. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 7, characterized in that, The objective weights are calculated based on the information entropy of each data point, and specifically include: A sample dataset is constructed using each data point in any grid within the urban ecological data / urban development data as a sample. Standardize the data of each sample in the sample dataset; Calculate the information entropy of each data point; The corresponding weights are determined based on the information entropy of each data point.

10. The method for calculating urban energy consumption level and urban ecology-development coupling degree based on nighttime light remote sensing according to claim 1, characterized in that, The calculation formulas for the energy consumption index, the ecological index, and the development index are as follows: Among them, U a This represents energy consumption indicators, where a1 and a2 represent total energy consumption and per capita energy consumption, respectively. These are the corresponding weights; U b Indicating ecological indicators, b i This represents the i-th data point in n urban ecological data points. For the corresponding weights; U c c represents a development indicator. i This represents the i-th data item in m city development data items. The corresponding weights; The formula for calculating the urban energy consumption level and the urban ecology-development coupling degree C is as follows:

Images