Multivariable regional energy structure medium and long term path prediction method and device

Abstract

The invention discloses a medium and long term path prediction method and device for a multivariable regional energy structure. The method comprises the following steps: setting a processing strategy, setting a target total energy consumption demand of a measurement and calculation region, and constructing a policy derived path according to the target total energy consumption demand; through a policy index strategy and a carbon pinch point calculation strategy, a 60-year target non-fossil primary energy demand proportion, a target non-fossil energy prediction usage amount and a fossil energy prediction usage amount of the measurement and calculation area are obtained through calculation; determining a policy derived path energy structure of 30-60 years through recursive analysis; step-by-step prediction is carried out through an autoregression prediction strategy of logarithm transformation to obtain slice annual classification fossil energy proportion data; step-by-step prediction is carried out through a Markov chain step-by-step prediction model, and slice annual classification non-fossil energy ratio data is obtained; and through iterative analysis, an algorithm external expansion path is obtained and structure prediction is carried out to obtain a final prediction path. The method can effectively predict the medium-and-long-term regional clean energy structure and consumption based on policy indexes and objective conditions.

Description

Technical Field

[0001] The present invention relates to the field of new energy and energy-saving technologies, and in particular to a method and device for predicting medium- and long-term paths of a multivariable regional energy structure. Background Art

[0002] At present, the clean energy structure at the regional level in my country is still in the development stage and has not shown an obvious trend; the factors affecting the development of clean energy are unclear, and the parameters of some algorithms are insufficiently selected to make accurate predictions. Some algorithms may have too many parameters of influencing factors and the data requirements are too high, making them difficult to promote. The prediction scope of existing studies is too large, and most of them remain at the provincial level, making it difficult to play a role in target planning for cities and districts within the province, resulting in difficulties in setting comprehensive goals, formulating strategies, and implementing them. At the same time, the prediction methods for the changes and development of regional energy structures and the specific forms of paths are still based on idealized methods such as staged growth rates, and the setting of stages and growth rates lacks data support and the optimization method of dynamic planning.

[0003] Therefore, how to design and develop a medium- and long-term path prediction method for multivariable regional energy structure has become an urgent problem to be solved. Summary of the invention

[0004] To this end, the present invention provides a method and device for predicting the medium- and long-term paths of regional energy structures with multiple variables, which can effectively predict the medium- and long-term regional clean energy structure and usage based on policy indicators and objective conditions. At the same time, it focuses on both the macro-policy and prediction algorithm levels, comprehensively considers the impact of policy routes and prediction routes on specific path forms, and uses a Markov chain prediction model with known probabilities to make more objective and accurate predictions on the clean energy structure of different categories.

[0005] In order to achieve the above object, the present invention provides the following technical solution: a method for predicting the medium- and long-term path of a multivariable regional energy structure, comprising:

[0006] According to the objective factors of the measurement area, by setting the processing strategy, the target total energy consumption demand of the measurement area in the next 30 and 60 years is set; according to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, unit GDP carbon emission data, 30-60 years of energy and carbon emission forecast data are carried out to build a policy derivation path;

[0007] Through the policy indicator strategy, the proportion of non-fossil primary energy demand in the measurement area in the next 60 years is calculated; through the carbon pinch calculation strategy, the predicted non-fossil energy consumption and fossil energy consumption in the measurement area in the next 60 years are calculated;

[0008] The recursive analysis weights are set according to the preferences of the decision-makers in the slice year, and the obtained data on the non-fossil primary energy demand ratio and the predicted non-fossil energy use and fossil energy use in the 60-year target area are recursively analyzed, and the energy structure of the 30-60-year policy-derived path is determined based on the comprehensive index;

[0009] Through the logarithmic transformation autoregressive prediction strategy, the data on the proportion of fossil energy by category in each slice year is gradually predicted;

[0010] According to the key influencing factor analysis strategy, the Markov chain step-by-step prediction model is used to gradually predict and obtain the data on the proportion of non-fossil energy by category in the slice year;

[0011] According to the energy endowment restrictions of various energy categories under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year, the algorithm expansion path is obtained by performing iterative analysis of step-by-step increase or decrease; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

[0012] As a preferred solution for a multivariable regional energy structure medium- and long-term path prediction method, in the process of setting processing strategies and calculating the target total energy consumption demand of the region in the next 30 and 60 years, the set processing strategies include a phased growth rate calculation strategy, a GDP-energy consumption elasticity coefficient calculation strategy, a per capita-total energy consumption prediction strategy, an industrial structure energy consumption calculation strategy and a China TIMES calculation model.

[0013] As a preferred solution for a medium- and long-term path prediction method of a multivariable regional energy structure, in the process of calculating the non-fossil primary energy demand ratio of the measurement area in 60 years through the policy indicator strategy, the policy indicator strategy is the clean energy proportion and unit planning indicator of the measurement area in the next 30-60 years;

[0014] In the process of calculating the 60-year target non-fossil energy forecast and fossil energy forecast for the measurement area through the carbon pinch calculation strategy, the calculation formula of the carbon pinch calculation strategy is:

[0015]

[0016] In the formula, S i is the supply of energy i; D is the total regional energy demand; F i The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO 2Emission constraint; Z is the amount of clean energy required to meet regional energy demand.

[0017] As a preferred solution for a multivariable medium- and long-term path prediction method for regional energy structure, in the process of determining the 30-60 year policy-derived path energy structure based on the comprehensive index, the comprehensive index calculation formula is:

[0018] Q=W·X=∑W i X i

[0019] In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. i is the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

[0020] As a preferred solution for a medium- and long-term path prediction method of a multivariable regional energy structure, in the process of obtaining the data on the proportion of non-fossil energy by category in the slice year through step-by-step prediction by the Markov chain step-by-step prediction model, the expression of the Markov chain step-by-step prediction model is:

[0021]

[0022] Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

[0023] The present invention also provides a medium- and long-term path prediction device for a multivariable regional energy structure, based on the above medium- and long-term path prediction method for a multivariable regional energy structure, comprising:

[0024] The policy derivative path construction module is used to set the target total energy consumption demand of the measurement area in the next 30 and 60 years by setting the processing strategy according to the objective factors of the measurement area; the energy and carbon emission forecast data for 30-60 years are carried out according to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, and the policy derivative path is constructed;

[0025] The target non-fossil energy and fossil energy consumption data prediction module is used to calculate the target non-fossil primary energy demand ratio of the measurement area in 60 years through policy indicator strategies; and calculate the predicted non-fossil energy consumption and fossil energy consumption of the measurement area in 60 years through the carbon pinch calculation strategy;

[0026] The 30-60-year policy derivative path energy structure determination module is used to set recursive analysis weights according to the preferences of decision makers in the slice year, recursively analyze the obtained 60-year target non-fossil primary energy demand ratio and non-fossil energy forecast and fossil energy forecast data of the measured area, and determine the 30-60-year policy derivative path energy structure based on the comprehensive index;

[0027] The slice-year-by-category fossil energy share prediction module is used to gradually predict and obtain the slice-year-by-category fossil energy share data through the logarithmic transformation autoregressive prediction strategy;

[0028] The slice annual category non-fossil energy proportion prediction module is used to obtain the slice annual category non-fossil energy proportion data through the Markov chain step-by-step prediction model according to the key influencing factor analysis strategy;

[0029] The final prediction path acquisition module is used to obtain the algorithm expansion path by performing iterative analysis of step-by-step increase or decrease based on the energy endowment restrictions of each energy category under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

[0030] As a preferred solution for a multivariable regional energy structure medium- and long-term path prediction device, in the policy-derived path construction module, in the process of setting the target total energy consumption demand of the region in the next 30 and 60 years by setting the processing strategy, the set processing strategy includes a phased growth rate calculation strategy, a GDP-energy consumption elasticity coefficient calculation strategy, a per capita-total energy consumption prediction strategy, an industrial structure energy consumption calculation strategy and a China TIMES calculation model.

[0031] As a preferred solution of a multivariable regional energy structure medium- and long-term path prediction device, in the target non-fossil energy and fossil energy consumption data prediction module, in the process of calculating the target non-fossil primary energy demand ratio of the measurement area in 60 years through the policy indicator strategy, the policy indicator strategy is the clean energy proportion and unit planning indicator of the measurement area corresponding to the next 30-60 years;

[0032] In the process of calculating the 60-year target non-fossil energy forecast and fossil energy forecast for the measurement area through the carbon pinch calculation strategy, the calculation formula of the carbon pinch calculation strategy is:

[0033]

[0034] In the formula, S i is the supply of energy i; D is the total regional energy demand; Fi The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO 2 Emission constraint; Z is the amount of clean energy required to meet regional energy demand.

[0035] As a preferred solution of a multivariable regional energy structure medium- and long-term path prediction device, in the 30-60 years policy-derived path energy structure determination module, in the process of determining the 30-60 years policy-derived path energy structure according to the comprehensive index, the comprehensive index calculation formula is:

[0036] Q=W·X=∑W i X i

[0037] In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. i is the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

[0038] As a preferred solution of a multivariable regional energy structure medium- and long-term path prediction device, in the slice-year category non-fossil energy proportion prediction module, in the process of obtaining the slice-year category non-fossil energy proportion data through step-by-step prediction by the Markov chain step-by-step prediction model, the expression of the Markov chain step-by-step prediction model is:

[0039]

[0040] Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

[0041] The present invention has the following advantages: according to the objective factors of the measurement area, by setting the processing strategy, the target total energy consumption demand of the measurement area in the next 30 and 60 years is set; according to the set target total energy consumption demand of the measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, unit GDP carbon emission data, 30-60 years of energy and carbon emission situation forecast data are carried out to construct a policy derivation path; through the policy indicator strategy, the 60-year target non-fossil primary energy demand ratio of the measurement area is calculated; through the carbon pinch calculation strategy, the 60-year target non-fossil energy forecast and fossil energy forecast of the measurement area are calculated; according to the preferences of the decision makers in the slice year, the recursive analysis weights are set, and the obtained 60-year target non-fossil primary energy demand ratio of the measurement area is calculated. The demand ratio and the predicted non-fossil energy usage are recursively analyzed with the predicted fossil energy usage data, and the energy structure of the policy-derived path for 30-60 years is determined based on the comprehensive index; the data on the share of fossil energy by category in the slice year is gradually predicted through the autoregressive prediction strategy with logarithmic transformation; the data on the share of non-fossil energy by category in the slice year is gradually predicted based on the key influencing factor analysis strategy through the Markov chain step-by-step prediction model; according to the energy endowment restrictions of each energy category under the autoregressive prediction situation and the carbon emission restrictions compared with the previous year, the algorithm expansion path is obtained through an iterative analysis of step-by-step increase or decrease; the structure is predicted through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

[0042] The present invention can effectively predict the medium- and long-term regional clean energy structure and usage based on policy indicators and objective conditions. At the same time, it can also make more objective and accurate predictions of the clean energy structure by category by taking into account the impact of policy routes and prediction routes on specific path forms, using a Markov chain prediction model with known probabilities. The present invention can also be used for multi-scale energy path and structure prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0044] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0045] Figure 1 A schematic flow chart of a method for predicting medium- and long-term paths of a multivariable regional energy structure provided in Example 1 of the present invention;

[0046] Figure 2 A schematic diagram of a specific prediction process of a method for predicting a medium- and long-term path of a multivariable regional energy structure provided in Example 1 of the present invention;

[0047] Figure 3 A schematic diagram of the China TIMES model framework in a method for predicting a medium- and long-term path of a multivariable regional energy structure provided in Example 1 of the present invention;

[0048] Figure 4 A schematic diagram of carbon pinch strategy calculation in a method for predicting medium- and long-term paths of a multivariable regional energy structure provided in Example 1 of the present invention;

[0049] Figure 5 A schematic diagram of a comprehensive index in a method for predicting a medium- and long-term path of a multivariable regional energy structure provided in Example 1 of the present invention;

[0050] Figure 6 A schematic diagram of an energy structure prediction process in a method for predicting a medium- and long-term path of a multivariable regional energy structure provided in Example 1 of the present invention;

[0051] Figure 7 This is a schematic diagram of the architecture of a multivariable regional energy structure medium- and long-term path prediction device provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0052] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Example 1

[0054] See also Figure 1 and Figure 2Embodiment 1 of the present invention provides a method for predicting medium- and long-term paths of a multivariable regional energy structure, comprising the following steps:

[0055] S1. According to the objective factors of the measurement area, by setting the processing strategy, set the target total energy consumption demand of the measurement area in the next 30 and 60 years; according to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, unit GDP carbon emission data, 30-60 years of energy and carbon emission forecast data are carried out to build a policy derivative path;

[0056] S2. Calculate the proportion of non-fossil primary energy demand in the measurement area in the next 60 years through policy indicator strategies; calculate the predicted non-fossil energy consumption and fossil energy consumption in the measurement area in the next 60 years through carbon pinch calculation strategies;

[0057] S3. Set recursive analysis weights according to the preferences of decision makers in the slice year, recursively analyze the obtained 60-year target non-fossil primary energy demand ratio and non-fossil energy forecast and fossil energy forecast data of the measured area, and determine the energy structure of the 30-60-year policy derivative path based on the comprehensive index;

[0058] S4. Through the autoregressive prediction strategy of logarithmic transformation, gradually predict and obtain the data on the proportion of fossil energy by category in the slice year;

[0059] S5. According to the key influencing factor analysis strategy, the Markov chain step-by-step prediction model is used to gradually predict and obtain the data on the proportion of non-fossil energy by category in the slice year;

[0060] S6. According to the energy endowment limit of each energy category under the autoregressive prediction scenario and the carbon emission limit compared with the previous year, the algorithm expansion path is obtained by performing iterative analysis of step-by-step increase or decrease; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated according to the policy-derived path to obtain the final prediction path.

[0061] In this embodiment, in step S1, according to the objective factors of the measurement area, by setting the processing strategy, the target total energy consumption demand of the measurement area in the next 30 years and 60 years is set; according to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, 30-60 years of energy and carbon emission forecast data are carried out to construct a policy derivation path;

[0062] Among them, the set processing strategies include phased growth rate calculation strategy, GDP-energy consumption elasticity coefficient calculation strategy, per capita-total energy consumption forecasting strategy, energy consumption calculation strategy by industrial structure and China TIMES calculation model.

[0063] Specifically,

[0064] 1) Strategy for calculating growth rate in stages:

[0065] You can choose 10 years as a cycle. For example, if the goal of achieving per capita GDP at the level of moderately developed countries is to be achieved in 2035, the growth rate must be maintained at 4.61%. For example, the EIA Energy Outlook (2017) proposes three development scenarios and predicts China's future economic growth rate, such as 3.3% from 2030 to 2040.

[0066] 2) GDP-energy consumption elasticity coefficient calculation strategy

[0067] E(t) 弹 =aG(t) b(t) +ε

[0068] In the formula, E 弹 is the elasticity coefficient method for predicting the total regional energy consumption; G is the regional GDP; a bε are constants; t is the slice year time.

[0069] 3) Per capita total energy consumption forecast strategy

[0070] E(t) 人均 =E pcap (t)N(t)

[0071] In the formula, E 人均 E is the total regional energy consumption predicted by the per capita elasticity coefficient method; pcap (t) is the per capita energy consumption in the region in slice year t; N(t) is the total population in the region in slice year t.

[0072] 4) Energy consumption estimation strategy based on industrial structure

[0073] The calculation is done by analyzing the average energy consumption intensity of the region's primary, secondary and tertiary industries and combining it with the proportion of each industry's output value.

[0074] 5) China Times calculation model

[0075] like Figure 3 As shown in the figure, first collect and obtain data related to social and economic development, including GDP, industrial structure, population and urbanization rate. The collected data is input into the energy service demand forecasting module, and demand forecast is carried out through dynamic material flow analysis, econometric method, industrial life cycle curve method, elasticity coefficient method, etc. After the forecast is completed, the data is input into the TIMES calculation model of the energy system module, and the target total energy consumption demand of the measurement area is obtained through analysis; the total energy consumption demand is input into the CO 2 Generate carbon emission pathways in the Emissions module.

[0076] In this embodiment, in step S2, the policy indicator strategy is used to calculate the 60-year target non-fossil primary energy demand ratio of the measurement area; the 60-year target non-fossil energy forecast and fossil energy forecast of the measurement area are calculated by the carbon pinch calculation strategy;

[0077] Specifically, the policy indicator method corresponds to specific planning indicators within the period of 30-60 years, including the proportion and unit of clean energy. For example, relevant policies stipulate that the proportion of clean energy at the national macro level will reach more than 80% in 60 years. The present invention is based on the relevant policy indicators corresponding to the measurement area.

[0078] The calculation formula of the carbon pinch calculation strategy is:

[0079]

[0080] In the formula, S i is the supply of energy i; D is the total regional energy demand; F i The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO 2 Emission constraint; Z is the amount of clean energy required to meet regional energy demand.

[0081] like Figure 4 As shown, according to the energy demand and CO 2 Emission limits draw a composite energy demand curve based on energy supply and CO 2 Draw the composite energy supply curve based on the emission level; secondly, superimpose the two composite curves in the same coordinate system, and move the energy supply curve horizontally to the right so that the energy supply curve intersects with the rightmost side of the energy demand curve. The intersection point is the carbon pinch point. The horizontal distance between the initial position of the energy supply curve and the position after the shift is the amount of clean energy that meets the regional emission limit.

[0082] In this embodiment, in step S3, the recursive analysis weights are set according to the preferences of the decision makers in the slice year, the obtained 60-year target non-fossil primary energy demand ratio and the non-fossil energy forecast and fossil energy forecast data of the measurement area are recursively analyzed, and the energy structure of the 30-60-year policy-derived path is determined according to the comprehensive index;

[0083] Specifically, weights are set based on the preferences of decision makers in different slice years. For example, for 30 and 35 years, the plan with the highest GDP total will be preferred, while for 60 years, the plan with the lowest overall carbon emissions will be preferred. Recursive analysis is then performed based on the comprehensive index to determine the energy structure of the 30-60 policy-derived path.

[0084] The calculation formula of the comprehensive index is:

[0085] Q=W·X=∑W i X i

[0086] In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. i is the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

[0087] like Figure 5 As shown in Figure 2, if the region prefers higher economic benefits and the secondary industry and fossil energy industry account for a high proportion, it will focus on higher total energy and lower energy consumption per unit GDP. When the weights are controlled to be 1, W is set subjectively. 1,3 >1 / 6,W 2,4,6 <1 / 6.

[0088] In this embodiment, in step S4, the data of fossil energy proportion of each category in each slice year is gradually predicted and obtained through the autoregressive prediction strategy of logarithmic transformation;

[0089] Specifically, the prediction of the proportion of fossil energy by category is based on the downward trend reflected in historical data, and the autoregressive prediction method with logarithmic transformation is used to gradually predict the proportion of fossil energy by category in the slice year.

[0090] Logarithmic changes in the proportion of each energy consumption structure:

[0091]

[0092] in:

[0093] Choose coal as the denominator and other types of fossil energy as the numerator:

[0094]

[0095] In the formula, p represents a component containing p components; t represents a time series; is the reference component (coal); For other ingredients.

[0096] Use the time series of a variable as the dependent variable series, and the time series of the same variable shifted several periods into the past as the independent variable series, analyze the correlation between a dependent variable series and another or more independent variable series, and establish a regression equation for prediction.

[0097]

[0098] Time t is determined by the previous X t-1 ...X t-p It is determined by the weighting of and the random disturbance term (white noise), where it is a p-order autoregressive model.

[0099]

[0100] Then use the inverse transformation to convert the log ratio back to the percentage.

[0101] In this embodiment, in step S5, according to the key influencing factor analysis strategy, the Markov chain step-by-step prediction model is used to step-by-step predict, and the data on the proportion of non-fossil energy by category in the slice year is obtained;

[0102] Specifically, based on the external information of classified influencing factors obtained by person correlation analysis, weights are obtained based on the random forest algorithm, and step-by-step prediction is performed based on the analysis of key influencing factors and the Markov chain.

[0103] First, we extract relevant factors that affect structural transformation and the growth rate of clean energy, such as the number of corresponding clean energy technology patents five years ago, the potential for resource development, and funding conditions, conduct Pearson correlation analysis, and screen important features.

[0104] The Pearson correlation coefficient r between two variables X and Y is defined as the quotient of the covariance between the two variables and the standard deviation:

[0105]

[0106] The closer the absolute value of r is to 1, the stronger the correlation is; the closer it is to 0, the weaker the correlation is.

[0107] Random forest solves the weight of important features. The random forest algorithm (RF) has a high prediction accuracy, can effectively analyze nonlinear, collinear and interactive data, and can give a variable importance score (VIM) while analyzing the data. Use feature importance to set the weight of each important feature.

[0108]

[0109] Where n tree is the number of trees; ER t t is the out-of-bag error rate on the original data of the tree; ER tj is the out-of-bag error rate of the tth tree after randomly permuting the value of feature j.

[0110] Prediction of changes in the proportion of different energy sources: a→s(n) as an example: compare the weighted feature importance scores of wind energy a and solar energy s. If the solar energy resource endowment is high, the current development ratio is small, and the solar energy capital flows are large, then P a→s (n)>0 and P s→a (n)=0.

[0111] Vector of primary energy consumption structure status

[0112] S(n)={s a (n),s s (n),s w (n),s g (n),s n (n),s b (n)} where s x (n) are the proportions of wind energy, solar energy, hydropower, geothermal energy, nuclear energy and biomass energy respectively:

[0113] Compare multiple parameters pairwise to obtain the conversion probability between each other and generate a Markov chain step-by-step prediction model:

[0114]

[0115] Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

[0116] Based on the first forecast structure, calculate the non-fossil energy consumption in 35 years: Add resource endowment restrictions: ∑E x <E x,max ; Gradually reduce the excess energy conversion ratio. Then predict the energy structure for 40 to 60 years.

[0117] In this embodiment, in step S6, based on the energy endowment restrictions of each energy category under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year, an algorithm expansion path is obtained by performing an iterative analysis of step-by-step increase or decrease; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

[0118] Specifically, Figure 6 As shown in the figure, the prediction of specific energy consumption by energy category is obtained by weighting the autoregressive prediction of typical energy categories with high autocorrelation coefficients and the overall non-fossil energy consumption. Based on the energy endowment restrictions of each energy category under the autoregressive prediction, the carbon emission restrictions compared with the previous year, etc., an iterative analysis of step-by-step increase or decrease is performed to obtain the algorithm expansion path; finally, the algorithm expansion path is used for structural prediction, and the total energy amount is calculated by referring to the policy-derived path to obtain the final prediction path.

[0119] In summary, the present invention sets the target total energy consumption demand of the measurement area in the next 30 and 60 years by setting a processing strategy according to the objective factors of the measurement area; predicts the energy and carbon emission situation for 30-60 years according to the set target total energy consumption demand of the measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, and constructs a policy derivation path; calculates the proportion of the 60-year target non-fossil primary energy demand in the measurement area through a policy indicator strategy; calculates the 60-year target non-fossil energy forecast and fossil energy forecast in the measurement area through a carbon pinch calculation strategy; sets recursive analysis weights according to the preferences of the decision makers in the slice year, and calculates the 60-year target non-fossil primary energy demand in the measurement area. The proportion and predicted non-fossil energy consumption are recursively analyzed with the predicted fossil energy consumption data, and the energy structure of the policy-derived path for 30-60 years is determined based on the comprehensive index; the data on the proportion of fossil energy by category in the slice year is gradually predicted through the autoregressive prediction strategy of logarithmic transformation; according to the key influencing factor analysis strategy, the data on the proportion of non-fossil energy by category in the slice year is gradually predicted through the Markov chain step-by-step prediction model; according to the energy endowment restrictions of each energy category under the autoregressive prediction situation and the carbon emission restrictions compared with the previous year, the algorithm expansion path is obtained through an iterative analysis of step-by-step increase or decrease; the structural prediction is carried out through the algorithm expansion path, and the total energy amount is calculated according to the policy-derived path to obtain the final prediction path.

[0120] The present invention can effectively predict the medium- and long-term regional clean energy structure and usage based on policy indicators and objective conditions. At the same time, it can also make more objective and accurate predictions of the clean energy structure by category by taking into account the impact of policy routes and prediction routes on specific path forms, using a Markov chain prediction model with known probabilities. The present invention can also be used for multi-scale energy path and structure prediction.

[0121] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0122] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0123] Example 2

[0124] See also Figure 7 Embodiment 2 of the present invention further provides a device for predicting medium- and long-term paths of a multivariable regional energy structure, comprising:

[0125] The policy derivative path construction module 001 is used to set the target total energy consumption demand of the measurement area in the next 30 and 60 years by setting the processing strategy according to the objective factors of the measurement area; and to predict the energy and carbon emission situation in the next 30-60 years according to the target total energy consumption demand of the measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, and construct the policy derivative path;

[0126] The target non-fossil energy and fossil energy consumption data prediction module 002 is used to calculate the target non-fossil primary energy demand ratio of the measurement area in 60 years through policy indicator strategies; and calculate the target non-fossil energy predicted consumption and fossil energy predicted consumption of the measurement area in 60 years through carbon pinch calculation strategies;

[0127] The 30-60-year policy derivative path energy structure determination module 003 is used to set the recursive analysis weight according to the preferences of the decision makers in the slice year, recursively analyze the obtained 60-year target non-fossil primary energy demand ratio and non-fossil energy forecast and fossil energy forecast data of the measured area, and determine the 30-60-year policy derivative path energy structure according to the comprehensive index;

[0128] The slice-year-by-category fossil energy share prediction module 004 is used to gradually predict and obtain the slice-year-by-category fossil energy share data through a logarithmic transformation autoregressive prediction strategy;

[0129] The slice annual category non-fossil energy proportion prediction module 005 is used to obtain the slice annual category non-fossil energy proportion data through the Markov chain step-by-step prediction model according to the key influencing factor analysis strategy;

[0130] The final prediction path acquisition module 006 is used to obtain the algorithm expansion path by performing iterative analysis of step-by-step increase or decrease based on the energy endowment restrictions of each energy category under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year; perform structural prediction through the algorithm expansion path, and calculate the total energy amount based on the policy-derived path to obtain the final prediction path.

[0131] In this embodiment, in the policy-derived path construction module 001, in the process of setting the target total energy consumption demand of the region in the next 30 and 60 years by setting the processing strategy, the set processing strategy includes a phased growth rate calculation strategy, a GDP-energy consumption elasticity coefficient calculation strategy, a per capita-total energy consumption forecast strategy, an industrial structure energy consumption calculation strategy and a China TIMES calculation model.

[0132] In this embodiment, in the target non-fossil energy and fossil energy consumption data prediction module 002, in the process of calculating the target non-fossil primary energy demand ratio of the measurement area in 60 years through the policy indicator strategy, the policy indicator strategy is the clean energy proportion and unit planning index of the measurement area corresponding to the next 30-60 years;

[0133] In the process of calculating the 60-year target non-fossil energy forecast and fossil energy forecast for the measurement area through the carbon pinch calculation strategy, the calculation formula of the carbon pinch calculation strategy is:

[0134]

[0135] In the formula, S i is the supply of energy i; D is the total regional energy demand; F i The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO2 emission constraint; Z is the amount of clean energy required to meet regional energy demand.

[0136] In this embodiment, in the 30-60 years policy derived path energy structure determination module 003, in the process of determining the 30-60 years policy derived path energy structure according to the comprehensive index, the comprehensive index calculation formula is:

[0137] Q=W·X=∑W i X i

[0138] In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. iis the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

[0139] In this embodiment, in the slice annual category non-fossil energy proportion prediction module 005, in the process of obtaining slice annual category non-fossil energy proportion data through step-by-step prediction by the Markov chain step-by-step prediction model, the expression of the Markov chain step-by-step prediction model is:

[0140]

[0141] Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

[0142] It should be noted that the information interaction, execution process and other contents between the modules of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the present application, and will not be repeated here.

[0143] Example 3

[0144] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code for a method for predicting a medium- and long-term path of a multivariable regional energy structure is stored. The program code includes instructions for executing a method for predicting a medium- and long-term path of a multivariable regional energy structure of embodiment 1 or any possible implementation thereof.

[0145] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0146] Example 4

[0147] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0148] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute a multivariable regional energy structure medium- and long-term path prediction method of embodiment 1 or any possible implementation thereof.

[0149] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0150] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0151] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0152] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A method for predicting medium- and long-term paths of multivariable regional energy structures, characterized in that: include: According to the objective factors of the measurement area, by setting the processing strategy, the target total energy consumption demand of the measurement area in the next 30 and 60 years is set; According to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, 30-60 years of energy and carbon emission forecast data are carried out to build a policy derivative path; Through the policy indicator strategy, the proportion of non-fossil primary energy demand in the measurement area in the next 60 years is calculated; through the carbon pinch calculation strategy, the predicted non-fossil energy consumption and fossil energy consumption in the measurement area in the next 60 years are calculated; The recursive analysis weights are set according to the preferences of the decision-makers in the slice year, and the obtained data on the non-fossil primary energy demand ratio and the predicted non-fossil energy use and fossil energy use in the 60-year target area are recursively analyzed, and the energy structure of the 30-60-year policy-derived path is determined based on the comprehensive index; Through the logarithmic transformation autoregressive prediction strategy, the data on the proportion of fossil energy by category in each slice year is gradually predicted; According to the key influencing factor analysis strategy, the Markov chain step-by-step prediction model is used to gradually predict and obtain the data on the proportion of non-fossil energy by category in the slice year; According to the energy endowment restrictions of various energy categories under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year, the algorithm expansion path is obtained by performing iterative analysis of step-by-step increase or decrease; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

2. A method for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 1, characterized in that: In the process of setting processing strategies and calculating the target total energy consumption demand of the region in the next 30 and 60 years, the set processing strategies include phased growth rate calculation strategy, GDP-energy consumption elasticity coefficient calculation strategy, per capita-total energy consumption forecast strategy, energy consumption calculation strategy by industrial structure and China TIMES calculation model.

3. A method for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 2, characterized in that: In the process of calculating the non-fossil primary energy demand ratio of the measurement area in 60 years through the policy indicator strategy, the policy indicator strategy is the clean energy ratio and unit planning index of the measurement area in the next 30-60 years; In the process of calculating the 60-year target non-fossil energy forecast and fossil energy forecast for the measurement area through the carbon pinch calculation strategy, the calculation formula of the carbon pinch calculation strategy is: In the formula, S i is the supply of energy i; D is the total regional energy demand; F i The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO2 emission constraint; Z is the amount of clean energy required to meet regional energy demand.

4. A method for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 3, characterized in that: In the process of determining the energy structure of the 30-60 years policy-derived path based on the comprehensive index, the calculation formula of the comprehensive index is: Q=W·X=∑W i X i In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. i is the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

5. A method for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 4, characterized in that: In the process of obtaining the data of non-fossil energy proportion of slices by category through step-by-step prediction using the Markov chain step-by-step prediction model, the expression of the Markov chain step-by-step prediction model is: Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

6. A device for predicting medium- and long-term paths of a multivariable regional energy structure, using a method for predicting medium- and long-term paths of a multivariable regional energy structure as claimed in any one of claims 1 to 5, characterized in that: include: The policy derivative path construction module is used to set the target total energy consumption demand of the measurement area in the next 30 and 60 years by setting the processing strategy according to the objective factors of the measurement area; the energy and carbon emission forecast data for 30-60 years are carried out according to the target total energy consumption demand of the set measurement area, as well as the historical per capita energy consumption data, unit GDP energy consumption data, and unit GDP carbon emission data, and the policy derivative path is constructed; The target non-fossil energy and fossil energy consumption data prediction module is used to calculate the target non-fossil primary energy demand ratio of the measurement area in 60 years through policy indicator strategies; and calculate the predicted non-fossil energy consumption and fossil energy consumption of the measurement area in 60 years through the carbon pinch calculation strategy; The 30-60-year policy derivative path energy structure determination module is used to set recursive analysis weights according to the preferences of decision makers in the slice year, recursively analyze the obtained 60-year target non-fossil primary energy demand ratio and non-fossil energy forecast and fossil energy forecast data of the measured area, and determine the 30-60-year policy derivative path energy structure based on the comprehensive index; The slice-year-by-category fossil energy share prediction module is used to gradually predict and obtain the slice-year-by-category fossil energy share data through the logarithmic transformation autoregressive prediction strategy; The slice annual category non-fossil energy proportion prediction module is used to obtain the slice annual category non-fossil energy proportion data through the Markov chain step-by-step prediction model according to the key influencing factor analysis strategy; The final prediction path acquisition module is used to obtain the algorithm expansion path by performing iterative analysis of step-by-step increase or decrease based on the energy endowment restrictions of each energy category under the autoregressive prediction scenario and the carbon emission restrictions compared with the previous year; structural prediction is performed through the algorithm expansion path, and the total energy amount is calculated based on the policy-derived path to obtain the final prediction path.

7. A device for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 6, characterized in that: In the policy-derivative path construction module, in the process of setting the target total energy consumption demand of the region in the next 30 and 60 years by setting the processing strategy, the set processing strategy includes the phased growth rate calculation strategy, the GDP-energy consumption elasticity coefficient calculation strategy, the per capita-total energy consumption forecast strategy, the energy consumption calculation strategy by industrial structure and the China TIMES calculation model.

8. A device for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 7, characterized in that: In the target non-fossil energy and fossil energy consumption data prediction module, in the process of calculating the target non-fossil primary energy demand ratio of the measurement area in 60 years through the policy indicator strategy, the policy indicator strategy is the clean energy proportion and unit planning indicator of the measurement area corresponding to the next 30-60 years; In the process of calculating the 60-year target non-fossil energy forecast and fossil energy forecast for the measurement area through the carbon pinch calculation strategy, the calculation formula of the carbon pinch calculation strategy is: In the formula, S i is the supply of energy i; D is the total regional energy demand; F i The supply of energy i to meet regional energy demand; W i is the remaining supply; C in is the regional energy demand emission factor; C out,i is the supply emission factor of energy source i; L is the regional CO2 emission constraint; Z is the amount of clean energy required to meet regional energy demand.

9. A device for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 8, characterized in that: In the 30-60 years policy derived path energy structure determination module, in the process of determining the 30-60 years policy derived path energy structure according to the comprehensive index, the comprehensive index calculation formula is: Q=W·X=∑W i X i In the formula, Q is the comprehensive index of each solution. The larger the Q value, the better the solution. i is the weight parameter, the sum of all weights is equal to 1; X i (i=1,2,3,4,5,6) are the total energy consumption in 2060, the degree of completion of the target of 80% non-fossil energy share in 2060, energy consumption per unit GDP, non-fossil energy consumption per unit GDP, per capita energy consumption, and per capita non-fossil energy consumption.

10. A device for predicting medium- and long-term paths of a multivariable regional energy structure according to claim 9, characterized in that: In the slice annual category non-fossil energy proportion prediction module, in the process of obtaining slice annual category non-fossil energy proportion data through step-by-step prediction by the Markov chain step-by-step prediction model, the expression of the Markov chain step-by-step prediction model is: Where P x→y (n) represents the probability of converting type x energy into type y energy; a is wind energy; s is solar energy; w is hydropower; g is geothermal energy; n is nuclear energy; and b is biomass energy.

Images