A zero-carbon park planning method based on full life cycle carbon neutrality

Abstract

The present invention relates to a zero-carbon park planning method based on full life cycle carbon neutrality, belongs to the technical field of carbon emission planning, and solves the problem that refined carbon control planning cannot be carried out in the prior art. The method includes: obtaining the historical carbon emissions of the building units in the park and the corresponding carbon emission related factor data; constructing a training sample set based on the historical carbon emissions of the building units and the corresponding carbon emission related factor data; constructing a time series prediction neural network model, and training the time series prediction neural network model based on the training sample set to obtain a carbon emission prediction model; predicting the carbon emissions of each building unit in the park based on the carbon emission prediction model, and determining the key controlled building units based on the predicted carbon emissions; and determining the planning improvement measures of the key controlled building units based on the sample data of the key controlled building units. Refined carbon control planning is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of low-carbon park planning, and in particular to a zero-carbon park planning method based on full life cycle carbon neutrality. Background Art

[0002] Carbon neutrality means that within a certain period of time, the total amount of greenhouse gas emissions such as carbon dioxide generated by human beings is completely absorbed or offset through afforestation, energy conservation and emission reduction, etc. As an important practical form of achieving carbon neutrality, zero-carbon parks have received widespread attention.

[0003] The zero-carbon park aims to achieve net zero greenhouse gas emissions in the entire park by integrating the energy, buildings, transportation and other systems in the park. Building units in the park are one of the important sources of carbon emissions, including implicit carbon emissions during the construction process and carbon emissions from energy consumption during operation.

[0004] When determining the control measures for building units within the park, there is a lack of targeted methods based on accurate data prediction. Usually a one-size-fits-all approach is adopted, or the control focus is determined based on a single factor such as the size of the building, without in-depth analysis of the carbon emission characteristics and potential of each building unit, and no refined carbon control planning can be carried out, resulting in poor results of control measures and difficulty in achieving the overall carbon neutrality goal of the park. Summary of the invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a zero-carbon park planning method based on full life cycle carbon neutrality, so as to solve the problem that the existing refined carbon control planning cannot be carried out.

[0006] On the one hand, an embodiment of the present invention provides a zero-carbon park planning method based on full life cycle carbon neutrality, comprising the following steps:

[0007] Obtain the historical carbon emissions of the building units in the park and the corresponding carbon emission related factor data; the carbon emission related factor data includes: the built environment data, energy consumption time series data and social economic time series data of the building units; construct a training sample set based on the historical carbon emissions of the building units and the corresponding carbon emission related factor data;

[0008] Constructing a time series prediction neural network model, and training the time series prediction neural network model based on a training sample set to obtain a carbon emission prediction model;

[0009] Predict the carbon emissions of each building unit in the park based on the carbon emissions prediction model, and determine the key control building units based on the predicted carbon emissions;

[0010] Determine planning improvement measures for key-controlled building units based on sample data of key-controlled building units.

[0011] Based on the further improvement of the above method, the carbon emissions of each building unit in the park are predicted based on the carbon emissions prediction model, and the key control building units are determined based on the predicted carbon emissions, including:

[0012] Obtaining built environment data of each building unit and energy consumption time series data of T time steps before the current moment and social economic time series data as input data to the carbon emission prediction model, and predicting the carbon emissions of each building unit in multiple time steps in the future;

[0013] Determine the carbon emission potential of each building unit based on the carbon emissions of each building unit at multiple future time steps and the ecological factor data of each building unit;

[0014] Determine whether each building unit is a key controlled building unit based on its carbon emission potential.

[0015] Based on the further improvement of the above method, the planning improvement measures for key controlled building units are determined based on the sample data of key controlled building units, including:

[0016] For each key controlled building unit, calculate the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label, determine the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label, and determine the corresponding planning improvement measures based on the correlation.

[0017] Based on the further improvement of the above method, the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label is calculated, and the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label is determined, including:

[0018] For each sample of the key controlled building unit, calculate the correlation between each dimension of the sample's energy consumption time series data and socio-economic time series data and the sample label;

[0019] For each dimension, the average value of the correlation between the time series data of the dimension and the sample label in each sample of the key controlled building unit is calculated to obtain the correlation between the time series data of the dimension and the sample label.

[0020] Based on the further improvement of the above method, a training sample set is constructed based on the historical carbon emissions of building units and the corresponding carbon emission related factor data, including:

[0021] For each building unit, multiple sliding windows of length T are used to slide in parallel on the energy consumption time series data, socio-economic time series data, and carbon emission time series data;

[0022] The energy consumption time series data, socio-economic time series data, carbon emission time series data, and built environment data of the building unit obtained at each sliding step are used as the input data of the sample. The carbon emission time series data of multiple time steps after the sliding window on the carbon emission time series data are used as the labels of the sample, constituting a sample.

[0023] Based on the further improvement of the above method, the time series prediction neural network model includes:

[0024] A first feature extraction module, used for extracting a first feature based on the built environment data of the sample;

[0025] A second feature extraction module, used for extracting a second feature based on the energy consumption time series data and the socio-economic time series data of the sample;

[0026] A third feature extraction module, used for calculating a third feature based on the carbon emission time series data in the input data of the sample;

[0027] A feature fusion module, used for fusing the first feature, the second feature and the third feature to obtain a fusion feature;

[0028] The prediction module is used to predict the label of the sample based on the fusion features.

[0029] Based on the further improvement of the above method, the second feature extraction module includes:

[0030] A periodic feature extraction module is used to extract periodic features of time series data based on sample energy consumption time series data and socio-economic time series data;

[0031] A deep feature extraction module, used to extract the deep features of each time step using a cycle unit based on the periodic features;

[0032] The feature weighting module is used to perform weighted fusion on the deep features of each time step based on the carbon emissions of each time step to obtain the second feature.

[0033] Based on the further improvement of the above method, the weight of the deep feature of each time step is calculated using the following formula:

[0034] ;in, represents the weight of the deep feature at the t-1th time step, represents the weight of the deep feature at the t-th time step, represents the carbon emissions at the tth time step, Represents the smoothing factor.

[0035] Based on the further improvement of the above method, the periodic feature extraction module extracts periodic features based on the energy consumption time series data and social and economic time series data of the sample in the following manner:

[0036] For each dimension of time series data, K periodic components of the time series data of the dimension are calculated, and the time series data of the dimension is converted into K two-dimensional data based on the K periodic components;

[0037] Based on the two-dimensional convolutional network, feature extraction is performed on each two-dimensional data to obtain K two-dimensional features;

[0038] Each two-dimensional feature is converted into a one-dimensional feature; based on the weights of the K periodic components, the K one-dimensional features are weightedly fused to obtain the periodic features corresponding to the time series data of each dimension.

[0039] Based on the further improvement of the above method, a fast Fourier transform is performed on the time series data of each dimension to obtain the period of the time series data of each dimension and the amplitude of each period;

[0040] Take K periods with larger amplitudes as the K period components of the time series data of this dimension.

[0041] Compared with the prior art, the present invention obtains the historical carbon emissions of the building units in the park and the corresponding carbon emission-related factor data, combines the built environment data, energy consumption data and social and economic data of the building units, and through comprehensive data integration, constructs a training sample set that can reflect the carbon emissions of the building units throughout their life cycle, thereby laying the foundation for accurate carbon emissions prediction; by constructing a time series prediction neural network model and training it with the integrated sample set, an accurate carbon emission prediction model is obtained, solving the problem that traditional prediction methods are difficult to handle complex carbon emission-related data relationships, improving the accuracy of prediction, and providing a reliable basis for carbon emission management in park planning. By predicting the carbon emissions of each building unit in the park, the key control building units are determined based on the predicted carbon emissions, thereby improving the previous situation of lack of accurate positioning of key control objects. By determining the planning improvement measures of the key control building units based on the sample data of the key control building units, the problem of formulating personalized planning improvement measures is solved, and the effectiveness of the control measures is improved to achieve the overall planning goal of the zero-carbon park.

[0042] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0044] Figure 1 This is a flow chart of a zero-carbon park planning method based on full life cycle carbon neutrality according to an embodiment of the present invention;

[0045] Figure 2 This is a block diagram of a zero-carbon park planning system based on full life cycle carbon neutrality according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0047] A specific embodiment of the present invention discloses a zero-carbon park planning method based on full life cycle carbon neutrality, such as Figure 1 As shown, the following steps are included:

[0048] S1. Obtain the historical carbon emissions of the building units in the park and the corresponding carbon emission related factor data; the carbon emission related factor data includes: the built environment data of the building units, the energy consumption time series data and the social and economic time series data; construct a training sample set based on the historical carbon emissions of the building units and the corresponding carbon emission related factor data;

[0049] S2. Constructing a time series prediction neural network model, and training the time series prediction neural network model based on a training sample set to obtain a carbon emission prediction model;

[0050] S3. Predict the carbon emissions of each building unit in the park based on the carbon emissions prediction model, and determine the key control building units based on the predicted carbon emissions;

[0051] S4. Determine planning improvement measures for key-controlled building units based on sample data of key-controlled building units.

[0052] It should be noted that the building units within the park include industrial buildings, office buildings, commercial buildings, residential buildings, cultural and sports buildings, and administrative buildings.

[0053] Data on carbon emission-related factors include built environment data of building units, energy consumption time series data, and socio-economic time series data.

[0054] Energy consumption data include consumption data of water, electricity, gas and chemical fuels.

[0055] The built environment data of a building unit includes the stage of the building unit's life cycle, land area, volume ratio, building type, land use nature, land use mix, building area, building shape coefficient, building height, and building orientation. Socioeconomic data includes population, economy, product output, etc.

[0056] The whole life cycle includes three stages: materialization, operation and disposal. It should be noted that within a unit of time, a building unit can only be in one stage.

[0057] Compared with the prior art, the zero-carbon park planning method based on full life cycle carbon neutrality provided by the embodiment of the present invention obtains the historical carbon emissions of the park building units and the corresponding carbon emission related factor data, combines the built environment data, energy consumption data and social and economic data of the building units, and through comprehensive data integration, constructs a training sample set that can reflect the carbon emissions of the building units throughout the life cycle, thereby laying the foundation for accurate carbon emissions prediction; by constructing a time series prediction neural network model and training it with the integrated sample set, an accurate carbon emission prediction model is obtained, which solves the problem that the traditional prediction method is difficult to handle complex carbon emission related data relationships, improves the accuracy of the prediction, and provides a reliable basis for carbon emission management in park planning. By predicting the carbon emissions of each building unit in the park, the key control building units are determined based on the predicted carbon emissions, thereby improving the previous situation of lack of accurate positioning of key control objects. By determining the planning improvement measures of the key control building units based on the sample data of the key control building units, the problem of formulating personalized planning improvement measures is solved, and the effectiveness of the control measures is improved to achieve the overall planning goal of the zero-carbon park.

[0058] The historical carbon emissions of each building unit are the carbon emissions calculated per unit time in history.

[0059] After obtaining the historical carbon emissions of the building unit and the corresponding carbon emission related factor data, a training sample set is constructed based on the historical carbon emissions of the building unit and the corresponding carbon emission related factor data, including:

[0060] For each building unit, multiple sliding windows of length T are used to slide in parallel on the energy consumption time series data, socio-economic time series data, and carbon emission time series data;

[0061] The energy consumption time series data, socio-economic time series data, carbon emission time series data, and built environment data of the building unit obtained at each sliding step are used as the input data of the sample. The carbon emission time series data of multiple time steps after the sliding window on the carbon emission time series data are used as the labels of the sample, constituting a sample.

[0062] During implementation, the built environment data of the building unit is numerical data, and the energy consumption, social economy, and carbon emissions are time series data, that is, time series data composed of energy consumption, social economy, and carbon emissions at each time step in history. Energy consumption time series data and social economy time series data are both multidimensional time series data. Therefore, when constructing samples, a sliding window is used to extract time series data to construct samples.

[0063] During implementation, it is assumed that the dimension of the time series data (energy consumption time series data and socio-economic time series data) in the carbon emission related factor data is C, and the total dimension is C+1 after adding the carbon emission time series data. For each building unit that has obtained the historical carbon emissions and the corresponding carbon emission related factor data, C+1 sliding windows of length T are used to slide in parallel on the C+1 dimensional time series data. The sliding step can be set to one time step or multiple time steps, which can be set according to the time series data length and calculation accuracy requirements.

[0064] For each sliding step, the time series data in the current window constitutes the multidimensional time series data of the sample. The multidimensional time series data and the built environment data of the building unit serve as the input data of the sample; the carbon emission time series data of multiple time steps after the current sliding window are used as the label of the sample, thereby obtaining a sample and constructing a training sample set.

[0065] A time series prediction neural network model is constructed, and the time series prediction neural network model is trained based on the training sample set to obtain a carbon emissions prediction model.

[0066] Specifically, the time series prediction neural network model includes:

[0067] A first feature extraction module, used for extracting a first feature based on the built environment data of the sample;

[0068] A second feature extraction module, used for extracting a second feature based on the energy consumption time series data and the socio-economic time series data of the sample;

[0069] A third feature extraction module, used for calculating a third feature based on the carbon emission time series data in the input data of the sample;

[0070] A feature fusion module, used for fusing the first feature, the second feature and the third feature to obtain a fusion feature;

[0071] The prediction module is used to predict the label of the sample based on the fusion features.

[0072] During implementation, since built environment data is numerical data, and energy consumption and socio-economic data are time series data, features need to be extracted separately. In addition, carbon emissions in the past may have an impact on carbon emissions in the future, and carbon emissions in the past time steps have a certain reference significance for carbon emissions in the future. Therefore, the features of carbon emissions data at past moments are extracted as part of the prediction features to predict future carbon emissions.

[0073] Therefore, the constructed time series prediction neural network model includes a first feature extraction module, a second feature extraction module and a third feature extraction module, which respectively extract the characteristics of the built environment data, the characteristics of the energy consumption time series data and the socio-economic time series data, and the corresponding carbon emissions characteristics.

[0074] During implementation, the first feature extraction module may adopt a one-dimensional convolutional network structure with a convolution kernel size of 1.

[0075] The second feature extraction module is used to extract features of time series data. Specifically, the second feature extraction module includes:

[0076] A periodic feature extraction module is used to extract periodic features of time series data based on sample energy consumption time series data and socio-economic time series data;

[0077] A deep feature extraction module, used to extract the deep features of each time step using a cycle unit based on the periodic features;

[0078] The feature weighting module is used to perform weighted fusion on the deep features of each time step based on the carbon emissions of each time step to obtain the second feature.

[0079] During implementation, since time series data is usually a superposition of multiple processes and has multi-periodic attributes, the periodic features of time series data are extracted through the periodic feature extraction module in order to more fully mine the features and improve the prediction accuracy.

[0080] Specifically, the periodic feature extraction module extracts periodic features based on the sample's energy consumption time series data and socio-economic time series data in the following manner:

[0081] For each dimension of time series data, K periodic components of the time series data of the dimension are calculated, and the time series data of the dimension is converted into K two-dimensional data based on the K periodic components;

[0082] Based on the two-dimensional convolutional network, feature extraction is performed on each two-dimensional data to obtain K two-dimensional features;

[0083] Each two-dimensional feature is converted into a one-dimensional feature; based on the weights of the K periodic components, the K one-dimensional features are weightedly fused to obtain the periodic features corresponding to the time series data of each dimension.

[0084] During implementation, a fast Fourier transform is performed on the time series data of each dimension to obtain the period of the time series data of each dimension and the amplitude of each period;

[0085] Take K periods with larger amplitudes as the K period components of the time series data of this dimension.

[0086] During implementation, the frequency components of the time series data of each dimension are calculated by fast Fourier transform, and the main K frequency components are taken according to the amplitude of the frequency components. According to the conversion relationship between the adopted frequency, frequency and period of the time series data, K period components can be obtained. That is, the period corresponding to the first K frequencies with larger amplitudes is taken as the K period components of this dimension. During implementation, the size of K is determined according to the length of the time series data and the calculation accuracy requirements.

[0087] During implementation, after obtaining K periodic components of each dimensional time series data, the dimensional time series data is converted into two-dimensional data based on each periodic component; in the two-dimensional data, the columns and rows of the two-dimensional data respectively reflect the time series changes within the cycle and between cycles.

[0088] It should be noted that if the length of the time series data is not a multiple of a periodic component, zeros are padded after the two-dimensional data corresponding to the periodic component to make the two-dimensional data structure complete. When the two-dimensional feature is converted to a one-dimensional feature later, the padded bits are first deleted and then converted to a one-dimensional feature.

[0089] Then, feature extraction can be performed on each two-dimensional data based on the two-dimensional convolutional network to obtain K two-dimensional features. During implementation, the two-dimensional convolutional network can use an existing network, such as an Inception network, to extract two-dimensional features. The size of the extracted two-dimensional features is the same as that of the input two-dimensional data.

[0090] After extracting the two-dimensional features, the two-dimensional features are converted into one-dimensional features, and the obtained one-dimensional features have the same length as the time series data of this dimension.

[0091] Since the time series data of this dimension has K periodic components, that is, K one-dimensional features are obtained, and each one-dimensional feature contributes differently to the data, therefore, the K one-dimensional features are weightedly fused based on the weights of the K periodic components to obtain the periodic features corresponding to the time series data of each dimension.

[0092] During implementation, the amplitude corresponding to each periodic component is used as the weight of the periodic component. The amplitude of each periodic component, that is, the amplitude of the frequency component corresponding to each periodic component, that is, the amplitude of each frequency obtained by fast Fourier transform.

[0093] During implementation, the following formula is used to calculate the periodic characteristics of the time series data of this dimension:

[0094] ;in, represents the periodic eigenvalue of the tth time step of the cth dimension time series data, Represents the amplitude corresponding to the kth period component of the cth dimension time series data, Represents the eigenvalue of the tth time step of the one-dimensional feature corresponding to the kth periodic component of the cth dimension time series data.

[0095] For example, the size of a sample of energy consumption time series data and socio-economic time series data is , T is the number of time steps, and C is the dimension. The periodic feature extraction module extracts features from each dimension separately, and the obtained periodic features are still time series data, that is, they are still size.

[0096] After obtaining the periodic features, the recurrent unit is used to extract the deep features of each time step. That is, the periodic features obtained at each time step are used as the input of a time step, and the recurrent unit is used to extract the deep features.

[0097] During implementation, the recurrent unit may adopt multiple layers of gated recurrent units, each layer of gated recurrent units is used to extract hidden features based on input data; the input of the first layer of gated recurrent units is the periodic features of each time step; the input of each layer of gated recurrent units except the first layer includes the periodic features of each time step and the hidden features of the previous time step; the last layer of gated recurrent units outputs the deep features of each time step.

[0098] Since each time step has different impacts on future moments, the deep features of each time step are weighted and fused based on the carbon emissions at each time step to obtain the second feature, so that the extracted features are more conducive to the prediction of carbon emissions at future moments.

[0099] When implemented, the weight of the deep features at each time step is calculated using the following formula:

[0100] ;in, represents the weight of the deep feature at the t-1th time step, represents the weight of the deep feature at the t-th time step, represents the carbon emissions at the tth time step, Represents the smoothing factor.

[0101] The smoothing factor determines the degree of reliance on past time steps. ,when When is larger, newer time steps will have a greater impact, and when When is smaller, earlier time steps will have a greater influence. The weight for the first time step can be preset.

[0102] During implementation, the third feature extraction module can adopt an existing recurrent neural network unit, such as a multi-layer gated recurrent unit.

[0103] During implementation, the feature fusion module concatenates the first feature, the second feature, and the third feature to obtain a fusion feature. The prediction module predicts carbon emissions based on the fusion feature. The prediction module can use a fully connected layer to output prediction values ​​for multiple time steps.

[0104] During implementation, the following formula is used to calculate the training loss of the time series prediction neural network model:

[0105] ;in, represents the mean square error loss, Represents shape loss.

[0106] By considering the mean square error loss and shape loss, the optimization direction of the model can be more effectively guided and the prediction accuracy can be improved.

[0107] Specifically, the shape loss is calculated using the following formula:

[0108] ;

[0109] in, Represents the distance loss between the label sequence and the prediction sequence, represents the principal component loss between the label sequence and the prediction sequence, Represents the correlation loss between the label sequence and the prediction sequence, , and Represents weight.

[0110] In implementation, if the predicted sequence has the same shape as the label sequence, the distance difference at each time step does not fluctuate much. Therefore, the following formula is used to calculate the distance loss between the label sequence and the predicted sequence:

[0111] ; Where N represents the number of samples in the current training batch, represents the length of the label sequence, represents the value of the tth element in the label sequence of the i-th sample, represents the value of the tth element in the prediction sequence of the i-th sample, Represents the value of the jth element in the label sequence of the i-th sample, represents the value of the jth element in the prediction sequence of the i-th sample, Represents the signed distance function.

[0112] If the predicted sequence has the same shape as the label sequence, the principal components of the predicted sequence and the label sequence are not much different. Therefore, the following formula is used to calculate the principal component loss between the label sequence and the predicted sequence:

[0113] ; Where N represents the number of samples in the current training batch, represents the number of principal components, represents the quantity of all ingredients, Indicates taking The jth principal component of Indicates taking The jth principal component of Indicates taking The kth non-principal component of represents the Lp norm, represents the label sequence of the i-th sample, Represents the predicted sequence of the i-th sample.

[0114] During implementation, the components of the sequence are obtained by fast Fourier transform to obtain the frequencies of the sequence, where the frequencies with larger amplitudes are the principal components and the others are non-principal components. The number of principal components can be determined according to the length of the sequence.

[0115] If the predicted sequence has the same shape as the label sequence, the difference between the correlation coefficient between the predicted sequence and the label sequence and the autocorrelation of the label sequence is very small. Therefore, the following formula is used to calculate the correlation loss between the label sequence and the predicted sequence:

[0116] ;in, represents the label sequence of the i-th sample, represents the predicted sequence of the i-th sample, represents the cross-correlation function, represents the Lp norm, and N represents the number of samples in the current training batch.

[0117] By considering distance fluctuations, principal component differences, and correlation differences, the similarity between the prediction results and labels is more fully considered, so that the model can predict carbon emissions more accurately.

[0118] When the model training reaches the number of iterations or meets the preset training loss accuracy, the training is stopped and the carbon emission prediction model is obtained.

[0119] Based on the carbon emission prediction model, the carbon emissions of each building unit in the park are predicted, and the key control building units are determined based on the predicted carbon emissions. Specifically, it includes:

[0120] S31, acquiring the built environment data of each building unit and the energy consumption time series data of T time steps before the current moment and the social economic time series data as input data and inputting them into the carbon emission prediction model to predict the carbon emissions of each building unit in multiple time steps in the future;

[0121] S32, determining the carbon emission potential of each building unit based on the average carbon emissions of each building unit in multiple future time steps and the ecological factor of each building unit;

[0122] S33. Determine whether each building unit is a key controlled building unit based on its carbon emission potential.

[0123] During implementation, we first obtain the built environment data of each building unit and the energy consumption time series data and socio-economic time series data of T time steps before the current moment as the input data of the model to predict the carbon emissions of each building unit in the park in multiple future time steps.

[0124] During implementation, dynamic icons and spatial heat maps can be used to display the carbon emissions of each building unit in the park.

[0125] The carbon emission potential of each building unit is then determined based on the carbon emissions of each building unit at multiple future time steps and the ecological factor data of each building unit.

[0126] The ecological factor data of the building unit include: green space type, green space area, and carbon sink coefficient. The carbon sink coefficient method is used to estimate the carbon absorption of the building unit during the predicted time period.

[0127] Based on the carbon emissions of each building unit at multiple time steps in the future and the ecological factor data of each building unit, the following formula is used to determine the carbon emission potential of each building unit: :

[0128] ;

[0129] ;in, represents the carbon emissions of the i-th building unit at the t-th time step predicted by the model, represents the number of time steps predicted by the model, represents the carbon absorption of the i-th building unit estimated based on the ecological factor data during the forecast period, represents the GDP value of the i-th construction unit in the forecast period, The carbon emission intensity target, that is, the carbon emission intensity target per unit of GDP, can be determined based on the low-carbon development plans of the industrial park or the superior unit.

[0130] When implemented, the GDP of the ith construction unit during the forecast period is an estimated value.

[0131] Key controlled building units shall be determined based on their carbon emission potential. During implementation, a threshold may be set in advance and building units whose calculated carbon emission potential exceeds the threshold shall be designated as key controlled building units.

[0132] After the key building units are identified, the planning improvement measures for the key controlled building units are determined based on the sample data of the key controlled building units, including:

[0133] For each key controlled building unit, calculate the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label, determine the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label, and determine the corresponding planning improvement measures based on the correlation.

[0134] Specifically, the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label is calculated, and the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label is determined, including:

[0135] For each sample of the key controlled building unit, calculate the correlation between each dimension of the sample's energy consumption time series data and socio-economic time series data and the sample label;

[0136] For each dimension, the average value of the correlation between the time series data of the dimension and the sample label in each sample of the key controlled building unit is calculated to obtain the correlation between the time series data of the dimension and the sample label.

[0137] During implementation, for each key controlled building unit, the dimension of the time series data with the greatest correlation with future carbon emissions is determined based on its sample, and corresponding planning improvement measures are formulated based on this dimension, so that the improvement measures are targeted and a zero-carbon park is achieved.

[0138] During implementation, for each sample of the key controlled building unit, the correlation between the energy consumption time series data and the socio-economic time series data of each dimension and the sample label is calculated. For example, the Pearson correlation can be calculated. The mean of the correlation corresponding to the time series data of this dimension of all samples of the key controlled building unit is the correlation value between the time series data of this dimension of the key controlled building unit and the label. The dimensions are sorted according to the correlation, and a preset number of dimensions with large correlations can be selected to determine the corresponding planning improvement measures for these dimensions.

[0139] For example, if the dimensions with greater correlation are water and electricity energy consumption, the corresponding planning measures can be improvement measures to reduce water and electricity energy consumption, thereby achieving a zero-carbon park.

[0140] A specific embodiment of the present invention discloses a zero-carbon park planning system based on full life cycle carbon neutrality, such as Figure 2 As shown, it includes the following modules:

[0141] A sample set construction module is used to obtain the historical carbon emissions of the building units in the park and the corresponding carbon emission related factor data; the carbon emission related factor data includes: the built environment data, energy consumption time series data and social economic time series data of the building units; based on the historical carbon emissions of the building units and the corresponding carbon emission related factor data, a training sample set is constructed;

[0142] A model training module is used to construct a time series prediction neural network model, and train the time series prediction neural network model based on a training sample set to obtain a carbon emission prediction model;

[0143] A key control building unit determination module is used to predict the carbon emissions of each building unit in the park based on the carbon emissions prediction model, and determine the key control building units based on the predicted carbon emissions;

[0144] The improvement measures determination module is used to determine the planning improvement measures of the key controlled building units based on the sample data of the key controlled building units.

[0145] The above method embodiments and system embodiments are based on the same principle, and their related parts can be used for reference, and can achieve the same technical effect. The specific implementation process refers to the above embodiment, which will not be repeated here.

[0146] A specific implementation of the present application discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of a zero-carbon park planning method based on full life cycle carbon neutrality are implemented.

[0147] A specific implementation of the present application discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a zero-carbon park planning method based on full life cycle carbon neutrality.

[0148] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0149] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A zero-carbon park planning method based on full life cycle carbon neutrality, characterized in that: The following steps are involved: Obtain historical carbon emissions of building units in the park and corresponding carbon emission-related factor data; The carbon emission related factor data include: built environment data, energy consumption time series data and social economic time series data of the building unit; a training sample set is constructed based on the historical carbon emissions of the building unit and the corresponding carbon emission related factor data; Constructing a time series prediction neural network model, and training the time series prediction neural network model based on a training sample set to obtain a carbon emission prediction model; Predict the carbon emissions of each building unit in the park based on the carbon emissions prediction model, and determine the key control building units based on the predicted carbon emissions; Determine planning improvement measures for key controlled building units based on sample data of key controlled building units; The training sample set is constructed based on the historical carbon emissions of the building unit and the corresponding carbon emission related factor data, including: For each building unit, multiple sliding windows of length T are used to slide in parallel on the energy consumption time series data, socio-economic time series data, and carbon emission time series data; The energy consumption time series data, social economic time series data, carbon emission time series data, and built environment data of the building unit obtained at each sliding step are used as the input data of the sample, and the carbon emission time series data of multiple time steps after the sliding window on the carbon emission time series data are used as the label of the sample, forming a sample; The time series prediction neural network model includes: A first feature extraction module, used for extracting a first feature based on the built environment data of the sample; A second feature extraction module, used for extracting a second feature based on the energy consumption time series data and the socio-economic time series data of the sample; A third feature extraction module, used for calculating a third feature based on the carbon emission time series data in the input data of the sample; A feature fusion module, used for fusing the first feature, the second feature and the third feature to obtain a fusion feature; Prediction module, used to predict the label of the sample based on the fusion features; The second feature extraction module comprises: A periodic feature extraction module is used to extract periodic features of time series data based on sample energy consumption time series data and socio-economic time series data; A deep feature extraction module, used to extract the deep features of each time step using a cycle unit based on the periodic features; The feature weighting module is used to perform weighted fusion on the deep features of each time step based on the carbon emissions of each time step to obtain the second feature.

2. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 1 is characterized in that: Based on the carbon emission prediction model, the carbon emissions of each building unit in the park are predicted, and the key control building units are determined based on the predicted carbon emissions, including: Obtaining built environment data of each building unit and energy consumption time series data of T time steps before the current moment and social economic time series data as input data to input into the carbon emission prediction model, and predicting the carbon emissions of each building unit in multiple time steps in the future; Determine the carbon emission potential of each building unit based on the carbon emissions of each building unit at multiple future time steps and the ecological factor data of each building unit; Determine whether each building unit is a key controlled building unit based on its carbon emission potential.

3. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 1 is characterized in that: Based on the sample data of key controlled building units, planning improvement measures for key controlled building units are determined, including: For each key controlled building unit, calculate the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label, determine the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label, and determine the corresponding planning improvement measures based on the correlation.

4. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 3 is characterized in that: Calculate the correlation between the time series data of each dimension in the carbon emission related factor data of each sample of the key controlled building unit and the sample label, and determine the correlation between the time series data of each dimension in the carbon emission related factor data and the sample label, including: For each sample of the key controlled building unit, calculate the correlation between each dimension of the sample's energy consumption time series data and socio-economic time series data and the sample label; For each dimension, the average value of the correlation between the time series data of the dimension and the sample label in each sample of the key controlled building unit is calculated to obtain the correlation between the time series data of the dimension and the sample label.

5. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 1 is characterized in that: The weight of the deep feature at each time step is calculated using the following formula: ； in, represents the weight of the deep feature at the t-1th time step, represents the weight of the deep feature at the t-th time step, represents the carbon emissions at the tth time step, Represents the smoothing factor.

6. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 1 is characterized in that: The periodic feature extraction module extracts periodic features based on the energy consumption time series data and socio-economic time series data of the sample in the following manner: For each dimension of time series data, K periodic components of the time series data of the dimension are calculated, and the time series data of the dimension is converted into K two-dimensional data based on the K periodic components; Based on the two-dimensional convolutional network, feature extraction is performed on each two-dimensional data to obtain K two-dimensional features; Each two-dimensional feature is converted into a one-dimensional feature; based on the weights of the K periodic components, the K one-dimensional features are weightedly fused to obtain the periodic features corresponding to the time series data of each dimension.

7. The zero-carbon park planning method based on full life cycle carbon neutrality according to claim 6 is characterized in that: Perform fast Fourier transform on the time series data of each dimension to obtain the period of the time series data of each dimension and the amplitude of each period; Take K periods with larger amplitudes as the K period components of the time series data of this dimension.

Images