An energy optimization system

Abstract

The present application relates to a kind of energy optimization systems, it includes energy demand prediction module, it is used for machine learning prediction method and traditional prediction method to carry out prediction processing to the basic data and energy statistical history data, obtain energy demand prediction value and carbon emission prediction value;And energy system optimization module, it is used for according to the energy demand prediction value and planning boundary parameter, with the minimum total amount of carbon emission, the minimum total cost of energy consumption as goal obtains the best energy structure, and with the minimum total cost of power installation as goal obtains power installation structure;The present application is predicted to energy demand by fusing ARIMA, regression model and traditional unit GDP energy consumption method, lays data foundation for high reliable subsequent optimization;While explicitly modeling and solving the multi-objective problem of cost minimization and carbon emission minimization, provide scientific, transparent decision-making basis for user, help to realize the win-win of economy and environmental protection.

Description

Technical Field

[0001] This invention relates to the field of energy optimization technology, and in particular to an energy optimization system. Background Technology

[0002] Traditional carbon neutrality data analysis service platforms typically include functional modules such as data access, governance and data fusion, monitoring and intelligent analysis, and dispatch and command, in order to achieve the national goal of "carbon peaking and carbon neutrality".

[0003] However, traditional carbon neutrality data analysis service platforms primarily monitor and analyze existing data, optimizing based on the current situation. Future data predictions rely on traditional methods such as energy consumption per unit of GDP. This type of energy system lacks the ability to accurately predict future data changes and cannot provide contingency plans for energy system optimization in response to such changes. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to overcome the defects or deficiencies of the prior art and provide an energy optimization system.

[0005] An energy optimization system includes a data management and preprocessing module, an energy demand forecasting module, an energy system optimization module, and a carbon emission assessment and visualization module;

[0006] The data management and preprocessing module is used to manage and maintain the basic data and historical energy statistics data of the industry energy consumption standard library, carbon emission factor library, power project cost library, energy calorific value converted to standard coal coefficient library, renewable energy resource development upper limit library, and planning boundary parameter library.

[0007] The energy demand forecasting module is used to process the basic data and historical energy statistics data using machine learning forecasting methods and traditional forecasting methods to obtain predicted energy demand and carbon emission values. The predicted energy demand includes the predicted total energy consumption S′ for the planning year. j Forecast value of total final energy consumption Z′ j Total electricity consumption forecast value K′ j Predicted maximum load value R′ for the whole society j And the predicted value M′ of the actual renewable energy responsibility absorption weight 1j ;

[0008] The energy system optimization module is used to obtain the optimal energy structure with the goal of minimizing total carbon emissions and total energy consumption costs, and to obtain the power generation structure with the goal of minimizing the total cost of power generation, based on the predicted energy demand and planning boundary parameters.

[0009] The carbon emission assessment and visualization module is used for carbon emission assessment optimization, scenario analysis, and result visualization.

[0010] In one embodiment, the energy demand forecasting module includes an energy consumption forecasting unit, which is used to obtain the predicted energy consumption value S′ for the target year. j Specifically, it includes the following steps:

[0011] S211: Obtain historical total energy consumption, GDP growth rate, and average energy intensity decline rate;

[0012] S212: Establish an ARIMA model of total energy consumption based on historical total energy consumption;

[0013] S213: The first predicted value of total energy consumption is obtained using the ARIMA model for total energy consumption;

[0014] S214: The unit GDP energy consumption method is used to process the historical total energy consumption, GDP growth rate and average energy intensity decline rate to obtain the second total energy consumption forecast.

[0015] Specifically, the second forecast value for total energy consumption was obtained using the energy consumption per unit of GDP method, as follows:

[0016] Second, the projected total energy consumption = historical total energy consumption × (1 + GDP growth rate - average energy intensity decline rate) t

[0017] Wherein, the total historical energy consumption is the total historical energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year.

[0018] S215: Calculate the weighted average of the first and second total energy consumption forecasts based on the preset weight W1 to obtain the total energy consumption forecast S′ for the planning year. j Specifically, it is expressed as: the predicted total energy consumption value S′ j = W1 × Forecast value of total first energy consumption + (1-W1) × Forecast value of total second energy consumption.

[0019] In one embodiment, the energy demand forecasting module further includes a final energy consumption forecasting unit, which is used to obtain the predicted final energy consumption value Z′ for the target year. j Specifically, it includes the following steps:

[0020] Step S22 includes the following sub-steps:

[0021] S221: Obtain historical total terminal energy consumption, GDP growth rate, and average energy intensity decline rate;

[0022] S222: Establish an ARIMA model for total terminal energy consumption based on historical total terminal energy consumption;

[0023] S223: The first predicted value of total final energy consumption is obtained by using the ARIMA model of total final energy consumption;

[0024] S224: The unit GDP energy consumption method is used to process the historical total final energy consumption, GDP growth rate, and average energy intensity decline rate to obtain the predicted value of the second final energy consumption; specifically, the predicted value of the second final energy consumption obtained by the unit GDP energy consumption method is: Predicted value of second final energy consumption = Historical total final energy consumption × (1 + GDP growth rate - Average energy intensity decline rate) t

[0025] Wherein, the total historical terminal energy consumption is the total historical terminal energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year.

[0026] S225: Calculate the weighted average of the first and second total terminal energy consumption forecasts based on the preset weight W2 to obtain the total terminal energy consumption forecast Z′ for the planning year. j ;

[0027] Specifically, it is represented as: the predicted value of total final energy consumption Z′ j = W2 × Forecast value of total energy consumption in the first terminal + (1-W2) × Forecast value of total energy consumption in the second terminal.

[0028] In one embodiment, the energy demand forecasting module further includes a total social electricity consumption forecasting unit, which is used to obtain the forecast value K′ of total social electricity consumption for the target year. j Specifically, it includes the following steps:

[0029] S231: Obtain historical total electricity consumption, historical regional GDP, historical resident population, GDP growth rate, and resident population growth rate;

[0030] S232: Based on historical total electricity consumption, historical regional GDP and historical resident population, an exponential, logarithmic and multinomial derivation was performed, and after data normalization and feature screening, a multiple regression model for total electricity consumption was established.

[0031] S233: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and the resident population growth rate, the predicted total electricity consumption K′ for the planning year is obtained using the aforementioned multiple regression model for total electricity consumption. j .

[0032] In one embodiment, the energy demand forecasting module further includes a total social maximum load forecasting unit, which is used to obtain the predicted value R′ of the total social maximum load for the target year. j Specifically, it includes the following steps:

[0033] S241: Obtain historical maximum load of the whole society, historical maximum utilization hours of the whole society, forecast of total electricity consumption, regional GDP, historical resident population, GDP growth rate and resident population growth rate;

[0034] S242: Based on historical maximum utilization hours of the whole society, historical regional GDP, and historical permanent resident population, exponential, logarithmic, and multinomial derivations were performed. After data normalization and feature screening, a multiple regression model for maximum utilization hours of the whole society was established.

[0035] S243: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and resident population growth rate, the predicted value of the maximum utilization hours of the whole society is obtained using the aforementioned multiple regression model of the maximum utilization hours of the whole society. Then, based on the formula: Predicted value of the first maximum load of the whole society = Predicted value of the maximum utilization hours of the whole society × Predicted value of the total electricity consumption of the whole society K′... j The first predicted value of the maximum load on the entire society was obtained.

[0036] S244: Based on historical maximum social load, historical regional GDP, and historical resident population, exponential, logarithmic, and multinomial derivations are performed. After data normalization and feature screening, a multiple regression model for maximum social load is established.

[0037] S245: After calculating the regional GDP and the number of permanent residents for the planning year using the GDP growth rate and the resident population growth rate, the second predicted value of the maximum social load is obtained by using the multivariate regression model of the maximum social load.

[0038] S246: Calculate the weighted average of the first and second predicted maximum social load values ​​based on the preset weight W4 to obtain the predicted maximum social load value R′ for the planning year. j ;

[0039] Specifically, it is represented as: the predicted maximum load of the entire society, R′ j = W4 × First predicted maximum load of the whole society + (1-W4) × Second predicted maximum load of the whole society.

[0040] In one embodiment, the energy demand forecasting module further includes an actual renewable energy responsibility absorption weight forecasting unit, which is used to obtain the actual renewable energy responsibility absorption weight forecast value M′ for the target year. 1j Specifically, it includes the following steps:

[0041] S251: Obtain the historical actual renewable energy liability absorption weight;

[0042] S252: Establish an ARIMA model for actual renewable energy responsibility absorption weight based on historical actual renewable energy responsibility absorption weight;

[0043] S253: Obtain the MAPE value of the Actual Renewable Energy Consumption Weight (ARIMA) model and compare the MAPE value with a set standard threshold. If the MAPE value is less than the set standard threshold, the ARIMA model passes the verification and proceed to step S254; otherwise, proceed to step S255.

[0044] S254: Obtain the predicted value M′ of the actual renewable energy consumption weight for the planning year based on the aforementioned ARIMA model of actual renewable energy consumption weight. 1j ;

[0045] S255: Obtain the production time trend variables and the predicted set of trend variables for the planning year, and establish a univariate linear regression model for the actual renewable energy consumption weight based on the production time trend variables. Use the univariate linear regression model for the actual renewable energy consumption weight to obtain the predicted value M′ of the actual renewable energy consumption weight for the planning year based on the predicted set of trend variables for the planning year. 1j .

[0046] In one embodiment, the energy system optimization module aims to obtain the optimal energy structure with the goals of minimizing total carbon emissions and total energy consumption costs, and includes the following steps:

[0047] SA10 obtains the projected total energy consumption value S′ for the planning year. j The proportion of coal-fired power in the planned annual power transfers (M′) j =1-M′ 1j ;;

[0048] SA20 sets a target of achieving the minimum total CO2 emissions in the planned year. j(min) and the minimum total cost of energy consumption. 1j(min) The objective function is as follows:

[0049] CO2 j(min) =EF1×A 1j +EF2×A 2j +EF3×A 3j +EF4×A5j ×M′ j

[0050] Cost 1j(min) =c1×A 1j +c2×A 2j +c3×A 3j +c4×(A 4j +A 5j )

[0051] Among them, A ij Let c1 be the optimized total consumption value of the i-th type of energy in the planning year, where i=1 represents the optimized total coal consumption value in the planning year, i=2 represents the optimized total oil consumption value in the planning year, i=3 represents the optimized total natural gas consumption value in the planning year, i=4 represents the optimized primary electricity consumption value in the planning year, and i=5 represents the optimized external power supply value in the planning year; c1 is the unit consumption price of coal, c2 is the unit consumption price of oil, c3 is the unit consumption price of natural gas, and c4 is the unit consumption price of electricity, which can be obtained through the Internet or defined by the user.

[0052] SA30 sets constraints, including total constraints, factual constraints, evaluation indicator constraints, and peak-reaching constraints.

[0053] The total energy consumption constraint condition is used to limit the sum of energy consumption of all types of energy sources to be equal to the total energy consumption forecast value S′ output by the energy demand forecasting module M2. j Specifically, it is expressed as:

[0054]

[0055] Among them O j Other energy consumption, which is the portion remaining after subtracting total coal consumption, total oil consumption, total natural gas consumption, primary electricity consumption, and externally imported electricity consumption from total energy consumption, changes proportionally to total energy consumption.

[0056] The factual constraints are as follows:

[0057] |A ij -A i(j-1) |≤μ1|A i(j-1) |

[0058] Among them, A i(j-1)Let represent the total consumption of the i-th energy source in the year preceding the planning year. Here, i=1 represents the total coal consumption in the year preceding the planning year, i=2 represents the total oil consumption in the year preceding the planning year, i=3 represents the total natural gas consumption in the year preceding the planning year, i=4 represents the total primary electricity consumption in the year preceding the planning year, and i=5 represents the total externally sourced electricity consumption in the year preceding the planning year. μ1 represents the maximum range of change in consumption between the planning year and the previous year. This value is input by the user, and in this application, it is set to 10%.

[0059] The constraints of the assessment indicators include the assessment of the cumulative decline rate of carbon intensity and the assessment of the proportion of non-fossil energy consumption, which are used to assess the percentage decrease in carbon emission intensity from the planning year to the base year and to assess the proportion of non-fossil energy consumption in total energy consumption, respectively.

[0060] The assessment of the cumulative decline rate of carbon intensity is specifically expressed as follows:

[0061] F j -F g ≤0

[0062] Among them, F j The cumulative rate of decrease in carbon intensity from the planning year to the baseline year; CO2 j For the planned year's carbon dioxide emissions; G′ j The GDP for the planning year is obtained from the planning boundary parameter library; CO20 is the carbon dioxide emissions for the baseline year, G0 is the GDP for the baseline year, and F is the GDP for the baseline year. g The target value for the cumulative rate of carbon emission reduction is obtained through the planning boundary parameter library; the base year is set by the user.

[0063] The assessment of the proportion of non-fossil energy consumption is specifically expressed as follows:

[0064] N j -N g ≥0

[0065] Where, N j To plan the annual non-fossil energy consumption share, S′ j N represents the projected total energy consumption for the planning year obtained through the energy demand forecasting module M2. g The target value for the proportion of non-fossil energy consumption is obtained through the planning boundary parameter library.

[0066] The peak emission constraint is used to limit the carbon emissions of a planning year to no higher than the previous year if the planning year falls after the peak emission year. Specifically, it is expressed as follows:

[0067] CO2 j ≤CO2 j-1

[0068] Among them, CO2 j-1 To plan the total carbon emissions for the year to come; the peak emission constraint conditions.

[0069] SA40 uses a commercial or open-source solver that supports large-scale linear or quadratic programming to solve the objective function set in step SA20 based on the constraints set in step SA30. If the solver fails, it introduces slack variables and penalty functions to handle the constraints; if it still fails, it uses a greedy algorithm as a fallback solution; finally, it outputs the optimized value A of the i-th type of energy consumption for the planning year. 1j The plan aims to achieve the minimum total carbon emissions and the minimum total energy consumption cost by the specified year.

[0070] In one embodiment, the energy system optimization module obtains the power generation structure with the objective of minimizing the total cost of power generation capacity, including the following steps:

[0071] SC10 obtains the predicted value R′ of the maximum load of the whole society in the planning year. j The projected total electricity consumption for the year is K′. j The projected total final energy consumption value Z′ for the planning year j ;

[0072] SC20 sets a minimum total cost for achieving power generation capacity in the planned year. 2j(min) The objective function is specifically expressed as:

[0073] Cost 2j(min) =∑ i max(B ij -B i(j-1) ,0)×b i

[0074] Among them, B ij For the i-th type of installed power capacity in the planned year, B i(j-1) To plan the installed capacity of the i-th type of power generation for the year to come, b i Let be the unit installed cost of the i-th type of electricity; where i = 1 represents coal power, i = 2 represents pumped storage, i = 3 represents gas power, i = 4 represents hydropower, i = 5 represents nuclear power, i = 6 represents wind power, i = 7 represents photovoltaic power, i = 8 represents new energy storage, and i = 9 represents biomass power.

[0075] SC30 sets constraints, including power balance constraints, energy balance constraints, peak shaving balance constraints, planning boundary constraints, and evaluation index constraints.

[0076] The power balance constraint condition is used to limit the sum of the total load of all types of power generating units and the total load transferred in from outside the region to be greater than or equal to the maximum load of the whole society, specifically expressed as follows:

[0077] ∑ i (B ij ×f i )+0.93∑f≥R′ j

[0078] Among them, f i is the output coefficient of the i-th type of electric generator set, which is obtained through the planning boundary parameter library; f is the generation load of the external transmission channel, which is obtained through the planning boundary parameter library;

[0079] The aforementioned power balance constraint condition is used to limit the sum of power generation from various energy sources and total power transferred in from outside the region to be greater than or equal to the total power consumption of the whole society, specifically expressed as follows:

[0080] ∑ i (B ij ×H ij )+A 5j ≥K′ j

[0081] ∑ i＝2,4,5,6,7,8,9 (B ij ×H ij )≥A 4j

[0082] Among them, H ij This represents the number of hours of power generation utilization for the i-th type in the planning year. When i=1, it represents coal power; when i=2, it represents pumped storage; when i=3, it represents gas power; when i=4, it represents hydropower; when i=5, it represents nuclear power; when i=6, it represents wind power; when i=7, it represents photovoltaic power; when i=8, it represents new energy storage; and when i=9, it represents biomass power.

[0083] The aforementioned peak-shaving balance constraint condition is used to limit the peak-shaving available capacity of each energy installed capacity type during the wet and dry seasons to the sum of the peak-shaving available capacity of externally supplied power and the transmission network loss, meaning the peak-shaving capacity should be greater than or equal to the required peak-shaving capacity. Specifically, it is expressed as follows:

[0084] ∑ i (B ij ×n i )+∑(f×n′)≥R′ j ×(1-ar)+D

[0085] Where, n idenoted as , where i = 1 represents coal-fired power, i = 2 represents pumped storage, i = 3 represents gas-fired power, i = 4 represents hydropower, i = 5 represents nuclear power, i = 6 represents wind power, i = 7 represents photovoltaic power, i = 8 represents new energy storage, and i = 9 represents biomass power; n′ represents the proportion of adjustable peak capacity in the power generation load of the external transmission channel, which is obtained through the planning boundary parameter library; a represents the demand-side response, which is a user input value; r represents the peak-valley difference rate, which is a user input value; and D represents the daily spinning reserve, which is a user input value.

[0086] The planning boundary constraints are used to limit the upper limit of installed capacity, and are specifically expressed as follows:

[0087]

[0088] in, This represents the upper limit of the installed capacity of the i-th type of power generator set, which is obtained through the planning boundary parameter library;

[0089] The constraints of the evaluation indicators include the assessment of the proportion of electricity in final energy consumption, which is used to study the proportion of electricity consumption in the final energy consumption structure, and is specifically expressed as follows:

[0090] L j =(∑ i (B ij ×H ij )+A 5j )×E

[0091]

[0092] R=D j -D g ≥0

[0093] Among them, L j Let E be the total electricity consumption for the planned year, and D be the standard coal equivalent of electricity. j Dg represents the proportion of electricity in final energy consumption for the planned year, and Dg is the target value for the proportion of electricity in final energy consumption.

[0094] SA40 employs a commercial or open-source solver that supports large-scale linear or quadratic programming to solve the objective function set in step SC20 based on the constraints set in step SC30. If the solver fails, slack variables and penalty functions are introduced to handle the constraints; if the solution still fails, a greedy algorithm is used as a fallback solution. The final output is the i-th type of installed power capacity B for the planned year. ij And the minimum total cost of power installation.

[0095] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0096] Figure 1 This is a schematic diagram of the module structure of the present invention;

[0097] Figure 2 This is a flowchart of the energy consumption forecasting method of the present invention;

[0098] Figure 3 This is a flowchart of the terminal energy consumption total prediction method of the present invention;

[0099] Figure 4 This is a flowchart of the method for predicting total electricity consumption in society according to the present invention;

[0100] Figure 5 This is a flowchart of the method for predicting the maximum load of the entire society according to the present invention;

[0101] Figure 6 This is a flowchart of the actual renewable energy responsibility absorption weight prediction method of the present invention;

[0102] Figure 7 This is a flowchart of the optimal energy structure acquisition method of the present invention;

[0103] Figure 8 This is a flowchart of the method for obtaining the optimal power installed capacity structure according to the present invention. Detailed Implementation

[0104] The present invention will now be described in detail with reference to the accompanying drawings.

[0105] like Figure 1 As shown, the energy optimization system of the present invention includes a data management and preprocessing module M1, an energy demand forecasting module M2, and an energy system optimization module M3.

[0106] The data management and preprocessing module M1 is used to execute step S10: to manage and maintain the basic data, historical energy statistics data and planning boundary parameters of the industry energy consumption standard library, carbon emission factor library, power project cost library, energy calorific value converted to standard coal coefficient library, renewable energy resource development upper limit library, and planning boundary parameter library.

[0107] The energy demand forecasting module M2 is used to execute step S20: using machine learning forecasting methods and traditional forecasting methods to process the basic data and historical energy statistics data to obtain energy demand forecast values; wherein, the energy demand forecast values ​​include the total energy consumption forecast value S′ for the planning year. j Forecast value of total final energy consumption Z′ j Total electricity consumption forecast value K′ jPredicted maximum load value R′ for the whole society j And the predicted value M′ of the actual renewable energy responsibility absorption weight 1j .

[0108] The energy demand forecasting module M2 includes an energy consumption total forecasting unit, a terminal energy consumption total forecasting unit, a total social electricity consumption forecasting unit, a total social maximum load forecasting unit, and an actual renewable energy responsibility absorption weight forecasting unit.

[0109] like Figure 2 As shown, the total energy consumption prediction unit is used to perform step S21:

[0110] Step S21 specifically includes the following sub-steps:

[0111] S211: Obtain historical total energy consumption, GDP growth rate, and average energy intensity decline rate;

[0112] S212: Establish an ARIMA model of total energy consumption based on historical total energy consumption;

[0113] S213: The first predicted value of total energy consumption is obtained using the ARIMA model for total energy consumption;

[0114] S214: The unit GDP energy consumption method is used to process the historical total energy consumption, GDP growth rate and average energy intensity decline rate to obtain the second total energy consumption forecast.

[0115] Specifically, the second forecast value for total energy consumption was obtained using the energy consumption per unit of GDP method, as follows:

[0116] Second, the projected total energy consumption = historical total energy consumption × (1 + GDP growth rate - average energy intensity decline rate) t

[0117] Wherein, the total historical energy consumption is the total historical energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year.

[0118] S215: Calculate the weighted average of the first and second total energy consumption forecasts based on the preset weight W1 to obtain the total energy consumption forecast S′ for the planning year. j Specifically, it is expressed as: the predicted total energy consumption value S′ j = W1 × Forecast value of total first energy consumption + (1-W1) × Forecast value of total second energy consumption.

[0119] like Figure 3As shown, the terminal energy consumption total prediction unit is used to execute step S22:

[0120] Step S22 includes the following sub-steps:

[0121] S221: Obtain historical total terminal energy consumption, GDP growth rate, and average energy intensity decline rate;

[0122] S222: Establish an ARIMA model for total terminal energy consumption based on historical total terminal energy consumption;

[0123] S223: The first predicted value of total final energy consumption is obtained by using the ARIMA model of total final energy consumption;

[0124] S224: The unit GDP energy consumption method is used to process the historical total final energy consumption, GDP growth rate, and average energy intensity decline rate to obtain the predicted value of the second final energy consumption; specifically, the predicted value of the second final energy consumption obtained by the unit GDP energy consumption method is: Predicted value of second final energy consumption = Historical total final energy consumption × (1 + GDP growth rate - Average energy intensity decline rate) t

[0125] Wherein, the total historical terminal energy consumption is the total historical terminal energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year.

[0126] S225: Calculate the weighted average of the first and second total terminal energy consumption forecasts based on the preset weight W2 to obtain the total terminal energy consumption forecast Z′ for the planning year. j ;

[0127] Specifically, it is represented as: the predicted value of total final energy consumption Z′ j = W2 × Forecast value of total energy consumption in the first terminal + (1-W2) × Forecast value of total energy consumption in the second terminal.

[0128] like Figure 4 As shown, the total social electricity consumption prediction unit is used to execute step S23:

[0129] Step S23 includes the following sub-steps:

[0130] S231: Obtain historical total electricity consumption, historical regional GDP, historical resident population, GDP growth rate, and resident population growth rate;

[0131] S232: Based on historical total electricity consumption, historical regional GDP and historical resident population, an exponential, logarithmic and multinomial derivation was performed, and after data normalization and feature screening, a multiple regression model for total electricity consumption was established.

[0132] S233: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and the resident population growth rate, the predicted total electricity consumption K′ for the planning year is obtained using the aforementioned multiple regression model for total electricity consumption. j .

[0133] like Figure 5 As shown, the maximum load prediction unit for the entire society is used to execute step S24:

[0134] Step S24 includes the following sub-steps:

[0135] S241: Obtain historical maximum load of the whole society, historical maximum utilization hours of the whole society, forecast of total electricity consumption, regional GDP, historical resident population, GDP growth rate and resident population growth rate;

[0136] S242: Based on historical maximum utilization hours of the whole society, historical regional GDP, and historical permanent resident population, exponential, logarithmic, and multinomial derivations were performed. After data normalization and feature screening, a multiple regression model for maximum utilization hours of the whole society was established.

[0137] S243: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and resident population growth rate, the predicted value of the maximum utilization hours of the whole society is obtained using the aforementioned multiple regression model of the maximum utilization hours of the whole society. Then, based on the formula: Predicted value of the first maximum load of the whole society = Predicted value of the maximum utilization hours of the whole society × Predicted value of the total electricity consumption of the whole society K′... j The first predicted value of the maximum load on the entire society was obtained.

[0138] S244: Based on historical maximum social load, historical regional GDP, and historical resident population, exponential, logarithmic, and multinomial derivations are performed. After data normalization and feature screening, a multiple regression model for maximum social load is established.

[0139] S245: After calculating the regional GDP and the number of permanent residents for the planning year using the GDP growth rate and the resident population growth rate, the second predicted value of the maximum social load is obtained by using the multivariate regression model of the maximum social load.

[0140] S246: Calculate the weighted average of the first and second predicted maximum social load values ​​based on the preset weight W4 to obtain the predicted maximum social load value R′ for the planning year. j ;

[0141] Specifically, it is represented as: the predicted maximum load of the entire society, R′ j = W4 × First predicted maximum load of the whole society + (1-W4) × Second predicted maximum load of the whole society.

[0142] like Figure 6 As shown, the actual renewable energy responsibility absorption weight prediction unit is used to perform step S25:

[0143] Step S25 includes the following sub-steps:

[0144] S251: Obtain the historical actual renewable energy liability absorption weight;

[0145] S252: Establish an ARIMA model for actual renewable energy responsibility absorption weight based on historical actual renewable energy responsibility absorption weight;

[0146] S253: Obtain the MAPE value of the Actual Renewable Energy Consumption Weight (ARIMA) model and compare the MAPE value with a set standard threshold. If the MAPE value is less than the set standard threshold, the ARIMA model passes the verification and proceed to step S254; otherwise, proceed to step S255.

[0147] S254: Obtain the predicted value M′ of the actual renewable energy consumption weight for the planning year based on the aforementioned ARIMA model of actual renewable energy consumption weight. 1j ;

[0148] S255: Obtain the production time trend variables and the predicted set of trend variables for the planning year, and establish a univariate linear regression model for the actual renewable energy consumption weight based on the production time trend variables. Use the univariate linear regression model for the actual renewable energy consumption weight to obtain the predicted value M′ of the actual renewable energy consumption weight for the planning year based on the predicted set of trend variables for the planning year. 1j .

[0149] The energy system optimization module M3 is used to obtain the optimal energy structure based on the predicted energy demand and planning boundary parameters, with the goal of minimizing total carbon emissions and total energy consumption costs.

[0150] Furthermore, it also includes the carbon emission assessment and visualization module M4, which is used for carbon emission assessment optimization, scenario analysis, and result visualization.

[0151] like Figure 7 As shown, the energy structure optimization of the energy system optimization module M3 includes the following steps:

[0152] SA10 obtains the projected total energy consumption value S′ for the planning year. jAnd the predicted value of the actual renewable energy responsibility absorption weight M′ for the planning year 1j and through M′ j =1-M′ 1j The proportion of coal-fired power in the planned year's external power transfer was calculated as M′. j ;

[0153] SA20 sets a target of achieving the minimum total CO2 emissions in the planned year. j(min) and the minimum total cost of energy consumption. 1j(min) The objective function is as follows:

[0154] CO2 j(min) =EF1×A 1j +EF2×A 2j +EF3×A 3j +EF4×A 5j ×M′ j

[0155] Cost 1j(min) =c1×A 1j +c2×A 2j +c3×A 3j +c4×(A 4j +A 5j )

[0156] Among them, A ij Let c1 be the optimized total consumption value of the i-th type of energy in the planning year, where i=1 represents the optimized total coal consumption value in the planning year, i=2 represents the optimized total oil consumption value in the planning year, i=3 represents the optimized total natural gas consumption value in the planning year, i=4 represents the optimized primary electricity consumption value in the planning year, and i=5 represents the optimized external power supply value in the planning year; c1 is the unit consumption price of coal, c2 is the unit consumption price of oil, c3 is the unit consumption price of natural gas, and c4 is the unit consumption price of electricity, which can be obtained through the Internet or defined by the user.

[0157] SA30 sets constraints, including total constraints, factual constraints, evaluation indicator constraints, and peak-reaching constraints.

[0158] The total energy consumption constraint condition is used to limit the sum of energy consumption of all types of energy sources to be equal to the total energy consumption forecast value S′ output by the energy demand forecasting module M2. j Specifically, it is expressed as:

[0159]

[0160] Among them O jOther energy consumption, which is the portion remaining after subtracting total coal consumption, total oil consumption, total natural gas consumption, primary electricity consumption, and externally imported electricity consumption from total energy consumption, changes proportionally to total energy consumption.

[0161] The factual constraints are as follows:

[0162] |A ij -A i(j-1) |≤μ1|A i(j-1) |

[0163] Among them, A i(j-1) Let represent the total consumption of the i-th energy source in the year preceding the planning year. Here, i=1 represents the total coal consumption in the year preceding the planning year, i=2 represents the total oil consumption in the year preceding the planning year, i=3 represents the total natural gas consumption in the year preceding the planning year, i=4 represents the total primary electricity consumption in the year preceding the planning year, and i=5 represents the total externally sourced electricity consumption in the year preceding the planning year. μ1 represents the maximum range of change in consumption between the planning year and the previous year. This value is input by the user, and in this application, it is set to 10%.

[0164] The constraints of the assessment indicators include the assessment of the cumulative decline rate of carbon intensity and the assessment of the proportion of non-fossil energy consumption, which are used to assess the percentage decrease in carbon emission intensity from the planning year to the base year and to assess the proportion of non-fossil energy consumption in total energy consumption, respectively.

[0165] The assessment of the cumulative decline rate of carbon intensity is specifically expressed as follows:

[0166] F j -F g ≤0

[0167] Among them, F j The cumulative rate of decrease in carbon intensity from the planning year to the baseline year; CO2 j For the planned year's carbon dioxide emissions; G′ j The GDP for the planning year is obtained from the planning boundary parameter library; CO20 is the carbon dioxide emissions for the baseline year, G0 is the GDP for the baseline year, and F is the GDP for the baseline year. g The target value for the cumulative rate of carbon emission reduction is obtained through the planning boundary parameter library; the base year is set by the user.

[0168] The assessment of the proportion of non-fossil energy consumption is specifically expressed as follows:

[0169] N j -N g ≥0

[0170] Where, N j To plan the annual non-fossil energy consumption share, S′ j N represents the projected total energy consumption for the planning year obtained through the energy demand forecasting module M2. g The target value for the proportion of non-fossil energy consumption is obtained through the planning boundary parameter library.

[0171] The peak emission constraint is used to limit the carbon emissions of a planning year to no higher than the previous year if the planning year falls after the peak emission year. Specifically, it is expressed as follows:

[0172] CO2 j ≤CO2 j-1

[0173] Among them, CO2 j-1 To plan the total carbon emissions for the year to come; the peak emission constraint conditions.

[0174] SA40 uses a commercial or open-source solver that supports large-scale linear or quadratic programming to solve the objective function set in step SA20 based on the constraints set in step SA30. If the solver fails, it introduces slack variables and penalty functions to handle the constraints; if it still fails, it uses a greedy algorithm as a fallback solution; finally, it outputs the optimized value A of the i-th type of energy consumption for the planning year. 1j The plan aims to achieve the minimum total carbon emissions and the minimum total energy consumption cost by the specified year.

[0175] like Figure 8 As shown, the power installed capacity structure optimization of the energy system optimization module M3 includes the following steps:

[0176] SC10 obtains the predicted value R′ of the maximum load of the whole society in the planning year. j The projected total electricity consumption for the year is K′. j The projected total final energy consumption value Z′ for the planning year j ;

[0177] SC20 sets a minimum total cost for achieving power generation capacity in the planned year. 2j(min) The objective function is specifically expressed as:

[0178] Cost 2j(min) =∑ i max(B ij -B i(j-1) ,0)×b i

[0179] Among them, B ij For the i-th type of installed power capacity in the planned year, B i(j-1) To plan the installed capacity of the i-th type of power generation for the year to come, b iLet be the unit installed cost of the i-th type of electricity; where i = 1 represents coal-fired power, i = 2 represents pumped storage, i = 3 represents gas-fired power, i = 4 represents hydropower, i = 5 represents nuclear power, i = 6 represents wind power, i = 7 represents photovoltaic power, i = 8 represents new energy storage, and i = 9 represents biomass power.

[0180] SC30 sets constraints, including power balance constraints, energy balance constraints, peak shaving balance constraints, planning boundary constraints, and evaluation index constraints.

[0181] The power balance constraint condition is used to limit the sum of the total load of all types of power generating units and the total load transferred in from outside the region to be greater than or equal to the maximum load of the whole society, specifically expressed as follows:

[0182] ∑ i (B ij ×f i )+0.93∑f≥R′ j

[0183] Among them, f i is the output coefficient of the i-th type of power generator, which is obtained through the planning boundary parameter library; f is the power generation load of the external transmission channel, which is obtained through the planning boundary parameter library.

[0184] The aforementioned power balance constraint condition is used to limit the sum of power generation from various energy sources and total power transferred in from outside the region to be greater than or equal to the total power consumption of the whole society, specifically expressed as follows:

[0185] ∑ i (B ij ×H ij )+A 5j ≥K′ j

[0186] ∑ i＝2,4,5,6,7,8,9 (B ij ×H ij )≥A 4j

[0187] Among them, H ij This represents the number of hours of power generation utilization for the i-th type in the planning year. When i=1, it represents coal power; when i=2, it represents pumped storage; when i=3, it represents gas power; when i=4, it represents hydropower; when i=5, it represents nuclear power; when i=6, it represents wind power; when i=7, it represents photovoltaic power; when i=8, it represents new energy storage; and when i=9, it represents biomass power.

[0188] The aforementioned peak-shaving balance constraint condition is used to limit the peak-shaving available capacity of each energy installed capacity type during the wet and dry seasons to the sum of the peak-shaving available capacity of externally supplied power and the transmission network loss, meaning the peak-shaving capacity should be greater than or equal to the required peak-shaving capacity. Specifically, it is expressed as follows:

[0189] ∑ i (B ij ×n i )+∑(f×n′)≥R′ j ×(1-ar)+D

[0190] Where, n i denoted as , where i = 1 represents coal-fired power, i = 2 represents pumped storage, i = 3 represents gas-fired power, i = 4 represents hydropower, i = 5 represents nuclear power, i = 6 represents wind power, i = 7 represents photovoltaic power, i = 8 represents new energy storage, and i = 9 represents biomass power; n′ represents the proportion of adjustable peak capacity in the power generation load of the external transmission channel, which is obtained through the planning boundary parameter library; a represents the demand-side response, which is a user input value; r represents the peak-valley difference rate, which is a user input value; and D represents the daily spinning reserve, which is a user input value.

[0191] The planning boundary constraints are used to limit the upper limit of installed capacity, and are specifically expressed as follows:

[0192]

[0193] in, This represents the upper limit of the installed capacity of the i-th type of power generator set, which is obtained through the planning boundary parameter library.

[0194] The constraints of the evaluation indicators include the assessment of the proportion of electricity in final energy consumption, which is used to study the proportion of electricity consumption in the final energy consumption structure, and is specifically expressed as follows:

[0195] L j =(∑ i (B ij ×H ij )+A 5j )×E

[0196]

[0197] R=D j -D g ≥0

[0198] Among them, L j Let E be the total electricity consumption for the planned year, and D be the standard coal equivalent of electricity. j Dg represents the proportion of electricity in final energy consumption for the planned year, and Dg is the target value for the proportion of electricity in final energy consumption.

[0199] SA40 employs a commercial or open-source solver that supports large-scale linear or quadratic programming to solve the objective function set in step SC20 based on the constraints set in step SC30. If the solver fails, slack variables and penalty functions are introduced to handle the constraints; if the solution still fails, a greedy algorithm is used as a fallback solution. The final output is the i-th type of installed power capacity B for the planned year. ij And the minimum total cost of power installation.

[0200] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0201] (1) The data management and preprocessing modules can seamlessly integrate heterogeneous energy consumption data at different scales, such as national, provincial, county and park levels, and build a unified analysis model to overcome the bottleneck of traditional platforms being unable to adapt to multi-level regional data.

[0202] (2) The energy demand forecasting module combines machine learning models with traditional forecasting methods, and the advantages of both complement each other to obtain more stable and more realistic comprehensive forecast results.

[0203] (3) The slack variables combined with the penalty function method in the energy system optimization module allow the model to handle situations where strict constraints cannot be met flexibly. The greedy algorithm's fallback mechanism provides a backup feasible solution when the main model fails, enhancing the algorithm's robustness. Simultaneously, this framework explicitly models and solves the multi-objective problem of minimizing cost and carbon emissions, exploring the Pareto optimal solution set, providing users with a scientific and transparent basis for decision-making, and helping to achieve a win-win situation for both economic efficiency and environmental protection.

[0204] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a,” “the,” and “the” used in the embodiments and claims of this application are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that, unless otherwise stated, “a plurality” means two or more; the terms “first,” “second,” “third,” etc., are used only to distinguish and not to describe a particular order or sequence, nor should they be construed as indicating or implying relative importance. The term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items. When the above description relates to drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of this application, those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0205] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. An energy optimization system, characterized in that: It includes a data management and preprocessing module, an energy demand forecasting module, an energy system optimization module, and a carbon emission assessment and visualization module; The data management and preprocessing module is used to manage and maintain the basic data and historical energy statistics data of the industry energy consumption standard library, carbon emission factor library, power project cost library, energy calorific value converted to standard coal coefficient library, renewable energy resource development upper limit library, and planning boundary parameter library. The energy demand forecasting module is used to process the basic data and historical energy statistics data using machine learning forecasting methods and traditional forecasting methods to obtain predicted energy demand and carbon emission values. The predicted energy demand includes the predicted total energy consumption for the planning year. Total terminal energy consumption forecast Total electricity consumption forecast Predicted maximum load value for the whole society And the predicted value of the actual renewable energy responsibility absorption weight ; The energy system optimization module is used to obtain the optimal energy structure with the objectives of minimizing total carbon emissions and total energy consumption costs, and to obtain the power generation structure with the objective of minimizing the total cost of power generation capacity, based on the predicted energy demand and planning boundary parameters. Specifically: Obtaining the optimal energy structure includes the following steps: SA10 obtains the total energy forecast for the planning year. The proportion of coal-fired power in the planned annual power transfers =1- ; SA20 sets a target of achieving the minimum total carbon emissions in the planned year. and the lowest total cost of energy consumption The objective function is as follows: in, Let be the optimized total consumption value of the i-th type of energy in the planning year, where i=1 represents the optimized total coal consumption value in the planning year, i=2 represents the optimized total oil consumption value in the planning year, i=3 represents the optimized total natural gas consumption value in the planning year, i=4 represents the optimized primary electricity consumption value in the planning year, and i=5 represents the optimized external power supply value in the planning year; c1 is the unit consumption price of coal, c2 is the unit consumption price of oil, c3 is the unit consumption price of natural gas, and c4 is the unit consumption price of electricity, which can be obtained through the Internet or defined by the user; SA30 sets constraints, including total constraints, factual constraints, evaluation indicator constraints, and peak-reaching constraints; The total energy consumption constraint is used to limit the sum of energy consumption of all types of energy sources to be equal to the total energy consumption forecast value output by the energy demand forecasting module M2. Specifically, it is expressed as: in Other energy consumption, which is the portion remaining after subtracting total coal consumption, total oil consumption, total natural gas consumption, primary electricity consumption, and externally imported electricity consumption from total energy consumption, changes proportionally to total energy consumption. ; The factual constraints are as follows: in, Let represent the total consumption of the i-th energy source in the year preceding the planning year. When i=1, it represents the total consumption of coal in the year preceding the planning year; when i=2, it represents the total consumption of oil in the year preceding the planning year; when i=3, it represents the total consumption of natural gas in the year preceding the planning year; when i=4, it represents the total consumption of primary electricity in the year preceding the planning year; and when i=5, it represents the total consumption of externally transferred electricity in the year preceding the planning year. This value, which is the maximum range of change in consumption between the planning year and the previous year, is entered by the user and is set to 10%. The constraints of the assessment indicators include the assessment of the cumulative decline rate of carbon intensity and the assessment of the proportion of non-fossil energy consumption, which are used to assess the percentage decline in the cumulative carbon emission intensity from the planning year to the base year and to assess the proportion of non-fossil energy consumption in total energy consumption, respectively. The assessment of the cumulative decline rate of carbon intensity is specifically expressed as follows: in, The cumulative rate of decrease in carbon intensity from the planning year to the baseline year; The carbon dioxide emissions for the planned year; The GDP for the planning year is obtained through the planning boundary parameter library; Carbon dioxide emissions for the baseline year. GDP in the base year The target value for the cumulative rate of carbon emission reduction is obtained through a planning boundary parameter library; the base year is set by the user. The assessment of the proportion of non-fossil energy consumption is specifically expressed as follows: in, To plan the proportion of non-fossil energy consumption in the annual plan, The total energy consumption forecast for the planning year is obtained through the energy demand forecasting module M2. The target value for the proportion of non-fossil energy consumption is obtained through a planning boundary parameter library; The peak emission constraint is used to limit the carbon emissions of a planning year to no higher than the previous year if the planning year falls after the peak emission year. Specifically, it is expressed as follows: in, To plan the total carbon emissions for the year ending in [year]; the aforementioned peak emission constraints; SA40 uses a commercial or open-source solver that supports large-scale linear or quadratic programming to solve the objective function set in step SA20 based on the constraints set in step SA30. If the solver fails, it introduces slack variables and penalty functions to handle the constraints; if it still fails, it uses a greedy algorithm as a fallback solution; finally, it outputs the optimized value of the i-th type of total energy consumption for the planning year. The plan aims to achieve the minimum total carbon emissions and the minimum total energy consumption cost by the designated year. The carbon emission assessment and visualization module is used for carbon emission assessment optimization, scenario analysis, and result visualization.

2. The energy optimization system according to claim 1, characterized in that: The energy demand forecasting module includes an energy consumption forecasting unit, which is used to obtain the predicted energy consumption value for the target year. Specifically, it includes the following steps: S211: Obtain historical total energy consumption, GDP growth rate, and average energy intensity decline rate; S212: Establish an ARIMA model of total energy consumption based on historical total energy consumption; S213: The first predicted value of total energy consumption is obtained using the ARIMA model for total energy consumption; S214: The historical total energy consumption, GDP growth rate, and average energy intensity decline rate are processed using the unit GDP energy consumption method to obtain the second total energy consumption forecast. Specifically, the second forecast value for total energy consumption was obtained using the energy consumption per unit of GDP method, as follows: Wherein, the total historical energy consumption is the total historical energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year; S215: Calculate the weighted average of the first and second total energy consumption forecasts based on the preset weight W1 to obtain the total energy consumption forecast for the planning year. Specifically, this is expressed as: the projected total energy consumption. .

3. The energy optimization system according to claim 2, characterized in that: The energy demand forecasting module also includes a final energy consumption forecasting unit, which is used to obtain the forecast value of final energy consumption for the target year. Specifically, it includes the following steps: Step S22 includes the following sub-steps: S221: Obtain historical total end-use energy consumption, GDP growth rate, and average energy intensity decline rate; S222: Establish an ARIMA model for total terminal energy consumption based on historical total terminal energy consumption; S223: The first predicted value of total final energy consumption is obtained by using the ARIMA model of total final energy consumption; S224: The historical total energy consumption, GDP growth rate, and average energy intensity decline rate are processed using the unit GDP energy consumption method to obtain the predicted value of total second-terminal energy consumption; specifically, the unit GDP energy consumption method is used to obtain the predicted value of total second-terminal energy consumption as follows: Wherein, the total historical terminal energy consumption is the total historical terminal energy consumption of the available base year, which can be obtained through historical energy statistics data; the GDP growth rate and the average energy intensity decline rate are available set values, which are obtained through the planning boundary parameter library, and t represents the number of years between the planning year and the base year; S225: Calculate the weighted average of the first and second total terminal energy consumption forecasts based on the preset weight W2 to obtain the total terminal energy consumption forecast for the planning year. ; Specifically, it is expressed as follows: .

4. The energy optimization system according to claim 3, characterized in that: The energy demand forecasting module also includes a total electricity consumption forecasting unit, which is used to obtain the forecast value of total electricity consumption for the target year. Specifically, it includes the following steps: S231: Obtain historical total electricity consumption, historical regional GDP, historical resident population, GDP growth rate, and resident population growth rate; S232: Based on historical total electricity consumption, historical regional GDP and historical resident population, an exponential, logarithmic and multinomial derivation was performed, and after data normalization and feature screening, a multiple regression model for total electricity consumption was established. S233: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and the resident population growth rate, the predicted total electricity consumption for the planning year is obtained using the aforementioned multiple regression model for total electricity consumption. .

5. The energy optimization system according to claim 4, characterized in that: The energy demand forecasting module also includes a total social maximum load forecasting unit, which is used to obtain the forecast value of the total social maximum load for the target year. Specifically, it includes the following steps: S241: Obtain historical maximum load of the whole society, historical maximum utilization hours of the whole society, forecast of total electricity consumption, regional GDP, historical resident population, GDP growth rate and resident population growth rate; S242: Based on historical maximum utilization hours of the whole society, historical regional GDP, and historical permanent resident population, exponential, logarithmic, and multinomial derivations were performed. After data normalization and feature screening, a multiple regression model for maximum utilization hours of the whole society was established. S243: After calculating the regional GDP and resident population for the planning year using the GDP growth rate and resident population growth rate, the predicted value of the maximum utilization hours of the whole society is obtained using the aforementioned multiple regression model for the maximum utilization hours of the whole society. Then, based on the formula: Predicted value of the first maximum load of the whole society = Predicted value of the maximum utilization hours of the whole society × Predicted value of the total electricity consumption of the whole society... The first predicted value of the maximum load on the entire society was obtained. S244: Based on historical maximum social load, historical regional GDP, and historical resident population, exponential, logarithmic, and multinomial derivations are performed. After data normalization and feature screening, a multiple regression model for maximum social load is established. S245: After calculating the regional GDP and the number of permanent residents for the planning year using the GDP growth rate and the resident population growth rate, the second predicted value of the maximum social load is obtained by using the multivariate regression model of the maximum social load. S246: Calculate the weighted average of the first and second predicted maximum social load values ​​based on the preset weight W4 to obtain the predicted maximum social load value for the planning year. ; Specifically, it is expressed as follows: .

6. The energy optimization system according to claim 5, characterized in that: The energy demand forecasting module also includes an actual renewable energy responsibility absorption weight forecasting unit, which is used to obtain the predicted value of the actual renewable energy responsibility absorption weight for the target year. Specifically, it includes the following steps: S251: Obtain the historical actual renewable energy liability absorption weight; S252: Establish an ARIMA model for actual renewable energy responsibility absorption weight based on historical actual renewable energy responsibility absorption weight; S253: Obtain the MAPE value of the Actual Renewable Energy Consumption Weight (ARIMA) model and compare the MAPE value with a set standard threshold. If the MAPE value is less than the set standard threshold, the ARIMA model passes the verification and proceed to step S254; otherwise, proceed to step S255. S254: Obtain the predicted value of the actual renewable energy consumption weight for the planning year based on the ARIMA model of actual renewable energy consumption weight. ; S255: Obtain the production time trend variable and the predicted set of trend variables for the planning year, and establish a univariate linear regression model for the actual renewable energy consumption weight based on the production time trend variable. Use the univariate linear regression model for the actual renewable energy consumption weight to obtain the predicted value of the actual renewable energy consumption weight for the planning year based on the predicted set of trend variables for the planning year. .

7. The energy optimization system according to claim 1, characterized in that, The energy system optimization module, which aims to minimize the total cost of installed power capacity, obtains the power capacity structure, including the following steps: SC10 obtains the predicted maximum load of the entire society in the planning year. Forecast of total electricity consumption for the planning year Projected total final energy consumption for the planning year ; SC20 sets the minimum total cost for achieving power generation capacity in the planned year. The objective function is specifically expressed as: in, For the i-th type of installed power capacity in the planned year, To plan the installed capacity of Category i power generation for the year to come, Let be the unit installed cost of the i-th type of electricity; where i=1 represents coal power, i=2 represents pumped storage, i=3 represents gas power, i=4 represents hydropower, i=5 represents nuclear power, i=6 represents wind power, i=7 represents photovoltaic power, i=8 represents new energy storage, and i=9 represents biomass power. SC30 sets constraints, including power balance constraints, electricity balance constraints, peak shaving balance constraints, planning boundary constraints, and evaluation index constraints. The power balance constraint condition is used to limit the sum of the total load of all types of power generating units and the total load transferred in from outside the region to be greater than or equal to the maximum load of the whole society, specifically expressed as follows: in, is the output coefficient of the i-th type of electric generator set, which is obtained through the planning boundary parameter library; The power generation load of the external power transmission channel is obtained through the planning boundary parameter library; The aforementioned power balance constraint condition is used to limit the sum of power generation from various energy sources and total power transferred in from outside the region to be greater than or equal to the total power consumption of the whole society, specifically expressed as follows: in, This represents the number of hours of power generation utilization for the i-th type in the planning year. When i=1, it represents coal power; when i=2, it represents pumped storage; when i=3, it represents gas power; when i=4, it represents hydropower; when i=5, it represents nuclear power; when i=6, it represents wind power; when i=7, it represents photovoltaic power; when i=8, it represents new energy storage; and when i=9, it represents biomass power. The aforementioned peak-shaving balance constraint condition is used to limit the peak-shaving available capacity of each energy installed capacity type during the wet and dry seasons to the sum of the peak-shaving available capacity of externally supplied power and the transmission network loss, meaning the peak-shaving capacity should be greater than or equal to the required peak-shaving capacity. Specifically, it is expressed as follows: in, Let be the peak-shaving coefficient for the i-th type of electricity, where i=1 represents coal-fired power, i=2 represents pumped storage, i=3 represents gas-fired power, i=4 represents hydropower, i=5 represents nuclear power, i=6 represents wind power, i=7 represents photovoltaic power, i=8 represents new energy storage, and i=9 represents biomass power. The proportion of adjustable peak capacity in the power generation load of external transmission channels is obtained through the planning boundary parameter library; is the demand-side response, which is a user input value; r is the peak-valley difference rate, which is a user input value; D is the daily spinning reserve, which is a user input value. The planning boundary constraints are used to limit the upper limit of installed capacity, and are specifically expressed as follows: in, This represents the upper limit of the installed capacity of the i-th type of power generator set, which is obtained through the planning boundary parameter library; The constraints of the evaluation indicators include the assessment of the proportion of electricity in final energy consumption, which is used to study the proportion of electricity consumption in the final energy consumption structure, and is specifically expressed as follows: in, To determine the total electricity consumption for the planned year, The standard coal equivalent coefficient for electricity. The target for the year is the proportion of electricity in final energy consumption. The target value for the proportion of electricity in final consumption.

Images