Public building carbon emission quota subentry splitting method, storage medium and equipment

Abstract

The invention discloses a public building carbon emission quota subentry splitting method, a storage medium and equipment, and aims to predict the month-by-month building total carbon emission based on big data, split the month-by-month building total carbon emission into the day-by-day carbon emission of each subentry, and realize public building carbon emission quota subentry prediction. By using the method of the invention, a total amount management and control target of each month-by-month carbon emission can be established, and daily tracking is carried out. On the basis, through reasonable early warning and alarm mechanism design, timely and accurate management control on the carbon emission of each sub-item can be realized.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission technology, and in particular to a method, storage medium and equipment for itemizing carbon emission quotas of public buildings. Background Art

[0002] Public buildings usually have large total carbon emissions, high carbon emission intensity per unit area, and complex energy usage characteristics due to their complex functions, large size, many energy-consuming equipment, rapid business changes, and complex distribution branches. They are a top priority for building carbon emission management.

[0003] The carbon emission quota of public buildings can usually be further broken down according to the building energy consumption items or equipment. Common items include: lighting system (including indoor lighting, outdoor lighting, landscape lighting, etc.); power system (including water supply and drainage pumps, elevator system, fire fans, smoke exhaust fans, etc.); parking system (including parking lot lighting and parking lot fans); HVAC system (including refrigeration main unit, refrigeration pump, cooling pump and cooling tower); and other systems (such as special equipment).

[0004] However, in actual applications, although public buildings understand the overall carbon emission control target, it is difficult to break down the overall target into individual items, and it is also difficult to break down an overall annual target into each day. Since building operation and maintenance personnel are not clear about the carbon emission constraints for each item and each day, they have no idea where to start during operation, which ultimately still easily leads to the overall carbon emissions of the building exceeding the standard.

[0005] Although the national standard provides guidance and constraint values ​​for carbon emissions of various types of public buildings, there are still two problems: First, the guidance and constraint values ​​of carbon emissions are statistical averages of various types of public buildings and cannot reflect the individual characteristics of buildings. Therefore, the "one-size-fits-all" energy consumption indicators cannot well guide the actual operation of each building.

[0006] Second, the standard provides the total carbon emission index for public buildings, but does not provide a method for splitting the carbon emission index into individual items, or into months and days. In current industry research, there is no relevant research on how to split the total carbon emission control target of buildings in more detail in terms of items and time. This also directly led to the general feedback from the industry that it is difficult to implement the national standard after its implementation, and that the goals and actions are easily out of touch.

[0007] The error back propagation algorithm (BP neural network algorithm) in the artificial neural network algorithm (ANN) is a multi-layer forward network with unidirectional propagation. The network is generally called a three-layer perception layer, namely: input layer, hidden layer and output layer. It is divided into two stages: forward and backward propagation. For each training sample, the calculation error of each layer of the network is performed once, and the error requirement is met after multiple calculations. This type of method is generally used in combination with other concepts. For example, in the process of establishing a GA-BP building carbon emission prediction model, a genetic algorithm is used to optimize the accident weights and thresholds of the BP neural network. The specific method is: determine the coding length, generate the initial population, fitness function, selection, crossover, mutation operation, calculate the individual fitness value, select the best value, and train the network. When establishing the PCA-BP building carbon emission prediction model, the prediction factors are selected first, and then the principal components are extracted, and the principal component analysis is performed using spss22 statistical software. Then the LM algorithm is used for training.

[0008] In practical applications, the BP neural network algorithm has the problem of local minimization. From a mathematical point of view, the traditional BP neural network is a local search optimization method. It solves a complex nonlinear problem. The network weights are gradually adjusted along the direction of local improvement, which will cause the algorithm to fall into local extreme values ​​and the weights converge to the local minimum point, resulting in network training failure. In addition, the BP neural network has problems such as slow convergence speed, different network structure selection, and sample dependence.

[0009] Support vector machine method (SVM) and support vector regression machine method (SVR). Support vector machine (SVM) is a relatively good method to realize the idea of ​​structural risk minimization. Its machine learning strategy is the principle of structural risk minimization. When making a carbon emission budget, the first choice is to select the prediction sample features. This technology currently selects the time series of historical carbon emissions of buildings, daily average temperature, direct sunlight, diffuse sunlight, and date attributes as sample features. Then the data is preprocessed, including abnormal data processing and data normalization. Then the Gaussian function is used as the kernel function of the support vector machine, and the parameters are selected by enumeration method, grid search method, genetic algorithm and particle swarm optimization algorithm. Then the cross-validation CV, including LOO-CV and K-CV, is used to determine the final parameters. Then the regression prediction results are evaluated by MAPE and RMSE. Finally, the error is verified by the carbon emission abnormality judgment standard. Support vector regression machine (SVR) mainly realizes linear regression by constructing a linear decision function in a high-dimensional space after dimensionality increase. When using e-insensitive function, its basis is mainly e-insensitive function and kernel function algorithm. Combined with other concepts, for example, KPCA kernel principal component analysis is used in KPCA-WLSSVM. Through nonlinear mapping, the original space is mapped to a high-dimensional feature space, and then principal component analysis is performed. The difference between the least squares support vector machine (LSSVM) and the standard support vector machine is that it defines the error norm as the loss function and changes the inequality constraint to an equality constraint. In GM-LSSVM, the GM grey model improves the prediction accuracy of the model by mining and sorting the original data. The steps are: perform a cumulative generation (1-AGO) on the original data to weaken the volatility and randomness of the random sequence. Establish a first-order time response differential equation, use the least squares method to find its parameters, and determine the time influence equation. Output the data and perform a cumulative generation (IAGO) to obtain the prediction result.

[0010] The SVM algorithm is difficult to implement for large-scale training samples, because SVM uses quadratic programming to solve the support vector, and solving the quadratic programming will involve the calculation of an m-order matrix (m is the number of samples). When the number m is large, the storage and calculation of the matrix will consume a lot of machine memory and computing time. The classic support vector machine algorithm only gives a two-class classification algorithm, but in the actual application of data mining, it is generally necessary to solve multi-class classification problems. For each high-dimensional space in this space, how to determine F, that is, the kernel function, there is no suitable method at present, so for general problems, SVM just converts the difficulty of the complexity of high-dimensional space into the difficulty of finding the kernel function. And even after the kernel function is determined, when solving the problem classification, it is required to solve the quadratic programming of the function, which requires a lot of storage space. This is also a problem of SVM. For the support vector regression machine (SVR), if the fitted mathematical model expresses a certain curve in the multidimensional space, the result obtained according to the e-insensitive function is the "e-pipeline" including the curve and the training point. Among all the sample points, only the part of the sample points distributed on the "pipe wall" determines the position of the pipeline. This part of the training samples is called "support vector". In order to adapt to the nonlinearity of the training sample set, the traditional fitting method usually adds higher-order terms after the linear equation. This method is indeed effective, but the additional adjustable parameters inevitably increase the risk of overfitting.

[0011] The time series analysis method conducts statistical research on the time series data of each variable. Based on the time series analysis method and the regression fitting method, a carbon emission data prediction method suitable for commercial buildings is obtained. The specific steps are to establish the original sequence and adjust it in season. The data can be tested for balance and randomness. Eviews 5 is used to analyze the data, determine the number of differences d, and perform ADF test on each difference sequence to establish a second-order difference sequence. The model is established and predicted for the sequence, and the sequence is fitted and predicted based on physical principles (the carbon emissions of the building, including electricity consumption, oil consumption, and gas consumption, are analyzed and predicted separately, which not only makes the method of establishing the time series model simpler, but also facilitates the actual building energy management and the implementation of building energy-saving measures), and the predicted values ​​of the building's monthly gas consumption and monthly oil consumption are corrected.

[0012] The time series prediction method has the defect of prediction error because it emphasizes time series and does not consider the influence of external factors. When encountering large changes in the outside world, there will often be large deviations. The time series prediction method is better for medium- and short-term predictions than for long-term predictions. In addition, the building carbon emission prediction method based on time series analysis is mainly aimed at commercial buildings with basically unchanged occupancy rates from month to month, and is not applicable to buildings with certain changes in occupancy rates from month to month. Moreover, the predicted carbon emissions of this method are the total carbon emissions of the building for the whole year, and cannot predict the sub-item carbon emissions of the building.

[0013] Statistical splitting method based on field surveys. At present, there are few studies on methods for splitting carbon emissions that do not rely on testing. Based on previous research, some researchers have proposed a simplified model suitable for office buildings that can maintain similar accuracy. The model first needs to establish a carbon emission calculation model for different equipment systems, and then combine it with a short-term survey, using a semi-empirical + semi-physical model mode, taking the key parameters that cannot be measured during the survey as the adjustment object, iteratively solving, and using the data of the building's total carbon emissions bill as the constraint value, so as to obtain the carbon emissions split results of each sub-item.

[0014] Taking the split lighting system carbon emissions split for office buildings as an example, four parameters need to be determined: ① The total lighting power of each area (based on the equipment list, or the lighting design power per unit area) area); ② the number of hours of use per day in each area (on-site survey); ③ the comprehensive simultaneous use coefficient of lighting in each area (different patterns on weekdays and weekends); ④ the energy saving rate of lighting control methods in the inner area (natural lighting design, automatic dimming design). In the survey, it is necessary to focus on collecting several parameters: lunch break time, overtime intensity (no overtime, slight overtime, medium-intensity overtime and high-intensity overtime) and maximum / minimum opening rate.

[0015] Taking the carbon emission split of the air conditioning system for office buildings as an example, its sub-item carbon emissions are composed of three parts: cold and heat sources, distribution systems, and terminal equipment. Its carbon emissions are calculated by "air conditioning load Q / air conditioning system energy efficiency index SCOP". The air conditioning load Q is estimated using the BIN method, and the air conditioning system efficiency index SCOP is simplified to the function regression calculation of the current load rate PLR. Using the total carbon emission bill of the building as the boundary condition, iterative calculation is performed, and the air conditioning carbon emissions with the smallest error between the sum of the sub-items and the total carbon emissions are finally obtained, which is the result of the sub-item split.

[0016] This method has three main disadvantages. First, it is only applicable to equipment that is continuously turned on and has a basically unchanged frequency of use (such as lighting), and cannot be applied to intermittently turned on equipment or variable frequency equipment, such as supply and exhaust fans, domestic water pumps, variable frequency fans and water pumps, etc. The "simultaneous use coefficient" of such equipment cannot be obtained through on-site surveys, or the estimated error will be very large. Second, it is still subject to on-site surveys. For sub-items that cannot enter the site (such as tenants in public buildings, special machine rooms, etc.), it is also impossible to obtain the parameters required for carbon emissions splitting. Although the sub-items that can be surveyed on-site simplify the on-site work projects, the workload is still large, and it is impossible to achieve low-cost promotion of hundreds or thousands of projects. Third, since this method uses actual carbon emissions as the boundary condition for splitting, it cannot give the "reasonable energy use range" of each sub-item, and can only be split based on the existing total carbon emissions. It is also applicable to buildings that have no sub-item metering data at all (in fact, there are very few large public buildings with zero metering now). For buildings that are now generally equipped with low-voltage sub-item metering systems, if there are metering data for some sub-items, it may provide boundary conditions for the splitting of carbon emissions of other sub-items, but this method does not explore this issue. Summary of the invention

[0017] In view of the deficiencies in the prior art, the present invention aims to provide a method, storage medium and device for itemizing carbon emission quotas for public buildings.

[0018] In order to achieve the above object, the present invention adopts the following technical solution: A method for splitting carbon emission quotas of public buildings into sub-items, characterized by comprising the following steps: S1. Forecast the daily carbon emissions of each item in the next 12 months: S1.1. Relevance judgment: Using historical daily carbon emissions of each building component and historical daily average outdoor dry-bulb temperature Calculate the correlation coefficient between the historical daily carbon emissions of each sub-item of the building and the outdoor temperature, where i represents each sub-item, j represents the jth day of each month, and n is the total number of sub-items; determine whether the historical daily carbon emissions of each sub-item are related to the outdoor temperature based on the correlation coefficient; if the historical daily carbon emissions of a sub-item have a high correlation with the outdoor temperature, proceed to the prediction calculation of step S1.2; if the historical daily carbon emissions of a sub-item have a low correlation with the outdoor temperature, proceed to the prediction calculation of step S1.3; S1.2. Climate-related daily carbon emissions forecast: S1.2.1. Data acquisition, including: 1) Calculate the daily cooling degree days CDD and the daily heating degree days HDD; 2) Date type, i.e. 1 to 7, representing Monday to Sunday; 3) Season type, including cooling season, heating season and transition season. Each season type has its day segment defined by the user; 4) Historical daily carbon emissions by item; The calculation process of daily cooling degree days CDD and daily heating degree days HDD is as follows: 1.1) If there is no predicted temperature data for the future month, use the historical outdoor temperature and calculate the daily cooling degree days CDD and the daily heating degree days HDD by following the steps 1.2)-1.6); If there is predicted temperature data for the future month, use the outdoor temperature of the future month and directly calculate the daily cooling degree days CDD and the daily heating degree days HDD based on the calculation formula in the following step 1.6); 1.2) Read historical annual temperature data and collect the maximum and minimum outdoor temperatures of each month and each day in all historical N years, m = 1~12, y = 1~N; 1.3) Calculate the monthly average, cumulative annual average and standard deviation of the historical maximum outdoor temperature and monthly minimum outdoor temperature as follows:

[0019]

[0020]

[0021] in, represents the average of the daily maximum outdoor temperature in the mth month of the yth year, It represents the maximum outdoor temperature on the Dth day of the mth month of the yth year. represents the average of the daily minimum outdoor temperature in the mth month of the yth year, represents the lowest outdoor temperature on the Dth day of the mth month of the yth year, It represents the average of the maximum outdoor temperature in month m in the N-year history. It represents the average of the minimum outdoor temperature in month m in the N-year history. represents the standard deviation of the maximum outdoor temperature in month m in historical N years, represents the standard deviation of the minimum outdoor temperature in month m in historical years N; d represents the total number of days in month m; 1.4) The standard coefficients of the highest and lowest outdoor temperatures in each month of the history are calculated through standardization, and the weighted sum is used to obtain the stability coefficient of each month of the year:

[0022]

[0023] in, The standard coefficient representing the maximum outdoor temperature in the mth month of the yth year, The standard coefficient representing the minimum outdoor temperature in the mth month of the yth year, represents the stability coefficient of the mth month of the yth year, and the weights of the maximum outdoor temperature and the minimum outdoor temperature are both 0.5; 1.5) For each month m, select the stability coefficient of the mth month of each year in the N historical years The daily outdoor temperature of the whole month with the minimum value is taken as the outdoor temperature data of the mth month of the typical meteorological year; 1.6) Calculate the daily cooling degree days CDD and daily heating degree days HDD according to the following formula:

[0024] in, is the maximum temperature on the jth day in a typical meteorological year, is the minimum temperature on the jth day in a typical meteorological year, is the preset threshold; S1.2.2. Use the date type, season type, historical daily carbon emissions of sub-items, daily cooling degree days CDD, and daily heating degree days HDD obtained in step S1.2.1 to train the prediction algorithm, where the feature values ​​can be selected based on experience, such as daily cooling degree days CDD and daily heating degree days HDD, date type, and season type, and the learning input sample is the historical daily carbon emissions data of sub-items collected in step S1.1; the trained prediction algorithm can be used to predict the daily carbon emissions data of the corresponding sub-items; S1.3. Calculation of climate-independent monthly forecast carbon emissions: S1.3.1. For the sub-items with low correlation with outdoor temperature identified in step S1.1, for the month m to be predicted, directly use the arithmetic mean and average change rate of the monthly carbon emissions of the corresponding months in all historical years to calculate the monthly predicted carbon emissions, and then proceed to step S1.4 to split the sub-item daily carbon emissions; the calculation of the monthly predicted carbon emissions is shown in the following formula:

[0025]

[0026]

[0027]

[0028] in, represents the monthly carbon emissions of the ith item in the mth month of the yth year in history, represents the average monthly carbon emissions of the ith item in the mth month of the Nth year in history, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the rate of change of the ith item in the mth month of the yth year, It represents the average change rate of the ith sub-item in all years in the mth month; S1.3.2. Breakdown of daily carbon emissions by climate-independent sub-item: For the monthly forecast carbon emissions calculated in step S1.3.1, the daily proportion coefficient is calculated using the sliding week coefficient method, and then the monthly forecast carbon emissions are split into daily forecast carbon emissions according to the daily proportion coefficient; only 7 days forward every day is a sliding week, and each sliding week includes the completed Monday to Sunday; the calculation process is shown in the following formula:

[0029]

[0030] Among them, i represents the i-th item, j represents the j-th day of each month, and w represents the w-th day among the 1st to 7th days in the sliding week with the j-th day as the first day. Indicates the sliding weekday ratio coefficient, Indicates the actual daily carbon emissions of the sub-item on the first day of the sliding week. Indicates the actual carbon emissions of each day in the historical sliding week. represents the daily ratio coefficient for the jth day of each month, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the sum of the daily proportion coefficients of the ith sub-item of the month. It represents the daily predicted carbon emissions of the jth day of the ith sub-item of the month; S2. Issue early warning and / or alarm based on the obtained daily carbon emissions by item.

[0031] Furthermore, in step S1.1, the calculation of the correlation coefficient between the daily carbon emissions of each building item and the outdoor temperature specifically includes:

[0032] in, Historical daily carbon emissions and historical outdoor temperatures The correlation coefficient of , They represent historical daily carbon emissions and historical outdoor temperature data respectively. When the correlation coefficient is in the range of [0.4,1], it means that the daily carbon emissions of the corresponding sub-item are positively correlated with the outdoor temperature; when the correlation coefficient is in the range of [-1,-0.4], it means that the daily carbon emissions of the corresponding sub-item are negatively correlated with the outdoor temperature; when the correlation coefficient is in the range of (-0.4,0.4), it means that the correlation between the daily carbon emissions of the corresponding sub-item and the outdoor temperature is low.

[0033] Furthermore, in step 1.6), , =18 .

[0034] Furthermore, in step S1.3.2, if there is no saved sliding week sub-item daily proportion coefficient, you need to wait for a week before you can use it, and the 7 sub-item daily proportion coefficients of the first 7-day cycle that includes a complete Monday to Sunday will be used as the basis; if there are more than 1 week of data saved, the most recent sliding week will be used as the basis; if there are bad numbers, the most recent sliding week data with complete good numbers will be used to recalculate the sliding week's daily proportion coefficient, and then calculate the sub-item daily predicted carbon emissions.

[0035] Furthermore, the specific process of early warning is as follows: 2.1.1. Calculate the risk rate: Calculate the simulated monthly carbon emissions by item as follows :

[0036] Among them, h represents the hth day, Indicates the sum of the actual daily carbon emissions of the sub-items in the past days. It represents the daily predicted carbon emissions of the item on that day. Indicates the actual carbon emissions of yesterday. It represents the daily forecast carbon emissions of yesterday. The sum of the daily predicted carbon emissions of the sub-items representing the days that have not occurred; Calculate the carbon emissions progress value as follows:

[0037] Indicates the monthly itemized carbon emissions budget; The risk rate is calculated as follows: Risk Rate

[0038] Among them, when When ; 2.1.2. According to the carbon emission progress calculated in step 2.1.1, the sensitivity is judged and the risk level is judged according to the risk rate.

[0039] Furthermore, in step S2, the specific process of alarming is as follows: 2.2.1. When the actual carbon emissions per item per month is greater than the budgeted carbon emissions per item per month, an alarm is generated; 2.2.2. Determine the severity of the alarm: The excess rate is calculated as follows: Exceeding standard rate

[0040] in, is the actual carbon emissions per month, Itemize carbon emissions budget for the month; The sensitivity of the alarm and the severity of the alarm are determined based on the excess rate.

[0041] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0042] The present invention also provides a computer device, comprising a processor and a memory, wherein the memory is used to store a computer program; and when the processor is used to execute the computer program, the above method is implemented.

[0043] The beneficial effect of the present invention is that the method of the present invention predicts the total carbon emissions of buildings on a monthly basis based on big data, and splits it into daily carbon emissions of each sub-item, thereby realizing the sub-item splitting of carbon emission quotas of public buildings. Using the method of the present invention, the total amount control target of each carbon emission on a monthly basis can be established and tracked daily. On this basis, through the reasonable design of early warning and alarm mechanisms, timely and accurate management and control of each sub-item carbon emission can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings: Figure 1 A logical schematic diagram of correlation determination in an embodiment of the present invention; Figure 2 A logical schematic diagram of climate-related daily carbon emissions prediction in an embodiment of the present invention; Figure 3 A logical schematic diagram of CDD and HDD calculation in an embodiment of the present invention; Figure 4 This is a logical schematic diagram of the SVM prediction algorithm training in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0046] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0047] The present invention will be further described below in conjunction with the accompanying drawings. It should be noted that this embodiment is based on the technical solution and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to this embodiment.

[0048] This embodiment provides a method for splitting carbon emission quotas of public buildings into sub-items, including the following steps: S1. Forecast the daily carbon emissions of each item in the next 12 months: S1.1. Relevance judgment: Using historical daily carbon emissions of each building component and historical daily average outdoor dry-bulb temperature The correlation coefficients between the historical daily carbon emissions of each building sub-item and the outdoor temperature were calculated respectively, where i represents each sub-item, j represents the jth day of each month, and n is the total number of sub-items.

[0049] In this embodiment, the classic Pearson correlation coefficient in statistics can be used (PearsonCorrelation Coefficient), as shown in formula (1).

[0050] (1) in, Historical daily carbon emissions and historical outdoor temperatures The correlation coefficient of , Represent the historical daily carbon emissions and historical outdoor temperature data respectively. The value of the correlation coefficient is between [-1, 1], 0 represents irrelevant, (0, 1] represents positive correlation, and [-1, 0) represents negative correlation. At present, the prior art does not give a threshold for determining whether the carbon emissions of a building are correlated with the outdoor temperature. The inventors have found through statistical analysis of a large amount of measured data that when the threshold of the correlation coefficient is between ±0.4, the correlation between daily carbon emissions and outdoor temperature can be effectively determined.

[0051] Specifically, in the present invention, the correlation coefficient is divided into high correlation and low correlation, where high correlation includes positive correlation and negative correlation. When the correlation coefficient is in the range of [0.4, 1], it means that the daily carbon emissions of the corresponding sub-item are positively correlated with the outdoor temperature; when the correlation coefficient is in the range of [-1, -0.4], it means that the daily carbon emissions of the corresponding sub-item are negatively correlated with the outdoor temperature; when the correlation coefficient is in the range of (-0.4, 0.4), it means that the correlation between the daily carbon emissions of the corresponding sub-item and the outdoor temperature is low.

[0052] If the historical daily carbon emissions of a certain sub-item are positively or negatively correlated with the outdoor temperature, the process proceeds to the prediction calculation of step S1.2; otherwise, the process proceeds to the prediction calculation of step S1.3.

[0053] S1.2、Forecast of daily carbon emissions by climate-related items, such as Figure 2 As shown: S1.2.1. Data acquisition, including: 1) Calculate the daily cooling degree days CDD and the daily heating degree days HDD; 2) Date type, i.e. 1 to 7, representing Monday to Sunday; 3) Season type, including cooling season, heating season and transition season. Each season type has its day segment defined by the user; 4) Historical daily carbon emissions by item; The calculation process of daily cooling degree days CDD and daily heating degree days HDD is as follows: Figure 3 As shown: 1.1) If there is no predicted temperature data for the future month, use the historical outdoor temperature and calculate the daily cooling degree days CDD and daily heating degree days HDD through the following steps 1.2)-1.6); if there is predicted temperature data for the future month, use the outdoor temperature of the future month and directly calculate the daily cooling degree days CDD and daily heating degree days HDD based on the calculation formula in the following step 1.6).

[0054] 1.2) Read historical annual temperature data and collect the maximum and minimum outdoor temperatures for each day of each month in all N historical years, m = 1~12, y = 1~N.

[0055] 1.3) Calculate the monthly average, cumulative annual average and standard deviation of the historical maximum outdoor temperature and monthly minimum outdoor temperature as follows: (2) (3) (4) in, represents the average of the daily maximum outdoor temperature in the mth month of the yth year, It represents the maximum outdoor temperature on the Dth day of the mth month of the yth year. represents the average of the daily minimum outdoor temperature in the mth month of the yth year, represents the lowest outdoor temperature on the Dth day of the mth month of the yth year, It represents the average of the maximum outdoor temperature in month m in the N-year history. It represents the average of the minimum outdoor temperature in month m in the N-year history. represents the standard deviation of the maximum outdoor temperature in month m in historical N years, represents the standard deviation of the minimum outdoor temperature in month m in historical years N; d represents the total number of days in month m.

[0056] 1.4) The standard coefficients of the highest and lowest outdoor temperatures in each month of the history are calculated through standardization, and the weighted sum is used to obtain the stability coefficient of each month of the year: (5) (6) in, The standard coefficient representing the maximum outdoor temperature in the mth month of the yth year, The standard coefficient representing the minimum outdoor temperature in the mth month of the yth year; It represents the stability coefficient of the mth month of the yth year, and the weights of the maximum outdoor temperature and the minimum outdoor temperature are both 0.5.

[0057] 1.5) For each month m, select the stability coefficient of the mth month of each year in the N historical years The daily outdoor temperature of the whole month with the minimum value is taken as the outdoor temperature data of the mth month of the typical meteorological year, that is, it is considered that in the historical year, the stability coefficient The outdoor temperature of the month with the minimum value is most representative of the corresponding month in all historical years. For example, for January, determine the stability coefficient for January in all historical years. The minimum value of the stability coefficient The daily outdoor temperature for the entire month of January corresponding to the minimum value is taken as the outdoor temperature data for January in a typical meteorological year.

[0058] 1.6) Calculate the daily cooling degree days CDD and daily heating degree days HDD according to the following formula:

[0059] in, is the maximum temperature on the jth day in a typical meteorological year, is the minimum temperature on the jth day in a typical meteorological year, is a preset threshold. In a preferred embodiment, , =18 .

[0060] S1.2.2. Use the date type, season type, historical daily carbon emissions of sub-items, daily cooling degree days CDD, and daily heating degree days HDD obtained in step S1.2.1 to train the prediction algorithm, where the characteristic values ​​can be selected based on experience, such as daily cooling degree days CDD and daily heating degree days HDD, date type, and season type, and the learning input sample is the historical daily carbon emissions data of sub-items collected in step S1.1; the trained prediction algorithm can be used to predict the daily carbon emissions data of the corresponding sub-items.

[0061] In the present invention, the daily carbon emission data of each item can be calculated by a prediction algorithm. The specific prediction algorithm is not the invention point of the present invention and will not be described in detail here. In a specific implementation, a support vector machine (SVM) prediction algorithm, decision tree, linear regression and other model prediction algorithms can be used to predict and calculate the daily carbon emission data of each item. Figure 4 In one embodiment shown, the date type, season type, historical daily carbon emissions by item, daily cooling degree days CDD, and daily heating degree days HDD obtained in step S1.2.1 can be used as model input parameters, and the future daily carbon emissions data by item can be predicted based on the model established on this basis.

[0062] S1.3. Climate-independent daily carbon emissions forecast: S1.3.1. For the sub-items with low correlation with outdoor temperature identified in step S1.1, for the month m to be predicted, directly use the arithmetic mean and average change rate of the monthly carbon emissions of the corresponding months in all historical years to calculate the monthly predicted carbon emissions, and then proceed to step S1.3.2 to split the sub-item daily carbon emissions; the calculation of the monthly predicted carbon emissions is shown in the following formula:

[0063]

[0064]

[0065]

[0066] in, represents the monthly carbon emissions of the ith item in the mth month of the yth year in history, represents the average monthly carbon emissions of the ith item in the mth month of the Nth year in history, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the rate of change of the ith item in the mth month of the yth year, It represents the average rate of change of the ith sub-item in all years in the mth month.

[0067] S1.3.2. Breakdown of daily carbon emissions by climate-independent sub-item: For the monthly forecast carbon emissions calculated in step S1.3.1, the daily proportion coefficient is calculated using the sliding week coefficient method, and then the monthly forecast carbon emissions are split into daily forecast carbon emissions according to the daily proportion coefficient; only 7 days forward every day is a sliding week, and each sliding week includes the completed Monday to Sunday; the calculation process is shown in the following formula: ] (12) (13) Among them, i represents the i-th item, j represents the j-th day of each month, and w represents the w-th day among the 1st to 7th days in the sliding week with the j-th day as the first day. Indicates the sliding weekday ratio coefficient, Indicates the actual daily carbon emissions of the sub-item on the first day of the sliding week. Indicates the actual carbon emissions of each day in the historical sliding week. represents the daily ratio coefficient for the jth day of each month, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the sum of the daily proportion coefficients of the ith sub-item of the month. It represents the daily predicted carbon emissions for the jth day of the ith sub-item of the month.

[0068] It should be noted that if there is no saved sliding weekly sub-item daily ratio coefficient (such as when the system is newly run), you need to wait a week before using it. The 7 sub-item daily ratio coefficients of the first 7 days (must include the complete Monday to Sunday) will be used as the basis.

[0069] If more than one week of data is saved, the most recent sliding week shall be used as the basis; if there are bad numbers, the most recent sliding week data with complete good numbers shall be used to recalculate the daily proportion coefficient of the sliding week, and then calculate the daily predicted carbon emissions of the sub-items.

[0070] In one embodiment, in step S1.3.2, if there is no saved sliding week sub-item daily proportional coefficient, it is necessary to wait for a week before using it, and the 7 sub-item daily proportional coefficients of the first 7-day cycle containing a complete Monday to Sunday shall prevail; if there are more than 1 week of data saved, the most recent sliding week shall prevail; if there are bad numbers, the most recent sliding week data with complete good numbers shall be used to recalculate the sliding week's daily proportional coefficient, and then calculate the sub-item daily predicted carbon emissions.

[0071] S2. Early warning and / or alarming are performed based on the obtained daily carbon emissions by item. It should be noted that the early warning and alarming methods of this step can also be used for early warning of total carbon emissions, by replacing the itemized carbon emissions with the total carbon emissions, or the total carbon emissions can be understood as a special itemized item.

[0072] In one embodiment, in step S2, the specific process of early warning is as follows: 2.1. Itemized carbon emissions warning: 2.1.1. Calculate the risk rate: Calculate the simulated monthly carbon emissions by item as follows :

[0073] Among them, h represents the hth day, Indicates the sum of the actual daily carbon emissions of the sub-items in the past days. It represents the daily predicted carbon emissions of the item on that day. Indicates the actual carbon emissions of yesterday. It represents the daily forecast carbon emissions of yesterday. Indicates the sum of the daily predicted carbon emissions for the sub-items for days that have not occurred.

[0074] Calculate the carbon emissions progress value as follows:

[0075] Indicates the monthly itemized carbon emissions budget; The risk rate is calculated as follows:

[0076] The following example further illustrates the calculation method of risk rate.

[0077] Assume that it is 10:20 on November 29, and the monthly budgeted carbon emissions for November is 5300kWh, then: The sum of the actual daily carbon emissions of the past days of the month, that is, the actual cumulative carbon emissions from 0:00 on November 1 to 23:59 on November 28, is 5000kWh.

[0078] Yesterday's actual carbon emissions and daily carbon emissions forecast If they are 200kWh and 190kWh respectively, then the rate of change of carbon emissions yesterday = 200 / 190 = 1.0526.

[0079] The days that did not occur in the month are November 29 (the current day has not yet passed, so it is calculated as a full day) and November 30. Assuming that the daily carbon emissions forecast for November 29 are = 180 kWh, daily carbon emissions forecast for November 30 =190kWh.

[0080] The total predicted carbon emissions for the days that have not occurred is: 180×1.0526+190×1.0526=389.47kWh The simulated monthly carbon emissions are as follows: 5000+389.47=5389.47kWh Carbon emissions progress value P = monthly simulated carbon emissions by item / monthly budgeted carbon emissions by item = 5389.462 / 5300 = 107.8%.

[0081] Risk rate = P-1 = 107.8%-1 = 7.8%.

[0082] 2.1.2. According to the carbon emission progress calculated in step 2.1.1, the sensitivity is judged and the risk level is judged according to the risk rate.

[0083] The sensitivity determination conditions set in this embodiment are shown in Table 1.

[0084] Table 1

[0085] In this embodiment, the sensitivity scale has five levels, which can be configured by the user in the foreground, but the percentage range corresponding to each sensitivity scale can only be modified in the background.

[0086] The risk level determination conditions set in this embodiment are shown in Table 2.

[0087] Table 2

[0088] Similarly, the risk rate percentage range corresponding to each risk level can only be modified in the background.

[0089] 2.2 Alarm: 2.2.1. When the actual carbon emissions of a sub-item month is greater than the predicted carbon emissions of a sub-item month, an alarm is generated; 2.2.2. Determine the severity of the alarm: The excess rate is calculated as follows:

[0090] in, is the actual carbon emissions per month, Itemize your carbon emissions budget for the month.

[0091] The sensitivity of the alarm and the severity of the alarm are determined based on the excess rate.

[0092] The sensitivity discrimination conditions of the alarm set in this embodiment are shown in Table 3, and the severity discrimination conditions of the alarm set in this embodiment are shown in Table 4.

[0093] Table 3

[0094] In this embodiment, the alarm sensitivity scale is configured with five levels, which can be adjusted and modified by the user in the foreground, and the percentage range corresponding to each sensitivity scale can be uniformly configured in the background.

[0095] Table 4

[0096] In this embodiment, the alarm severity scale is configured with five levels, and the percentage range corresponding to each level of the scale can be uniformly configured in the background.

[0097] The present invention also discloses a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and the computer program implements the above method when executed by a processor.

[0098] The present invention also discloses a computer device, comprising a processor and a memory, wherein the memory is used to store a computer program; and when the processor is used to execute the computer program, the above method is implemented.

[0099] In summary, the present invention provides a method, storage medium and device for splitting carbon emission quotas of public buildings, which predicts the total carbon emissions of buildings on a monthly basis based on big data, and splits it into daily carbon emissions of each sub-item, so as to realize the prediction of carbon emission quotas of public buildings. Using the method of the present invention, it is possible to establish a total amount control target for each carbon emission amount on a monthly basis and track it daily. On this basis, through the reasonable design of early warning and alarm mechanisms, timely and accurate management and control of each sub-item carbon emission amount can be achieved.

[0100] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

[0101] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0102] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for splitting carbon emission quotas of public buildings, characterized in that: The steps include: S1. Forecast the daily carbon emissions of each item in the next 12 months: S1.

1. Relevance judgment: Using historical daily carbon emissions of each building component and historical daily average outdoor dry-bulb temperature Calculate the correlation coefficient between the historical daily carbon emissions of each sub-item of the building and the outdoor temperature, where i represents each sub-item, j represents the jth day of each month, and n is the total number of sub-items; determine whether the historical daily carbon emissions of each sub-item are related to the outdoor temperature based on the correlation coefficient; if the historical daily carbon emissions of a sub-item have a high correlation with the outdoor temperature, proceed to the prediction calculation of step S1.2; if the historical daily carbon emissions of a sub-item have a low correlation with the outdoor temperature, proceed to the prediction calculation of step S1.3; S1.

2. Climate-related daily carbon emissions forecast: S1.2.

1. Data acquisition, including: 1) Calculate the daily cooling degree days CDD and the daily heating degree days HDD; 2) Date type, i.e. 1 to 7, representing Monday to Sunday; 3) Season type, including cooling season, heating season and transition season. Each season type has its day segment defined by the user; 4) Historical daily carbon emissions by item; The calculation process of daily cooling degree days CDD and daily heating degree days HDD is as follows: 1.1) If there is no predicted temperature data for the future month, use the historical outdoor temperature and calculate the daily cooling degree days CDD and the daily heating degree days HDD by following the steps 1.2)-1.6); If there is predicted temperature data for the future month, use the outdoor temperature of the future month and directly calculate the daily cooling degree days CDD and the daily heating degree days HDD based on the calculation formula in the following step 1.6); 1.2) Read historical annual temperature data and collect the maximum and minimum outdoor temperatures of each month and each day in all historical N years, m = 1~12, y = 1~N; 1.3) Calculate the monthly average, cumulative annual average and standard deviation of the historical maximum outdoor temperature and monthly minimum outdoor temperature as follows: in, represents the average of the daily maximum outdoor temperature in the mth month of the yth year, represents the maximum outdoor temperature on the Dth day of the mth month of the yth year, represents the average daily minimum outdoor temperature in the mth month of the yth year, represents the lowest outdoor temperature on the Dth day of the mth month of the yth year, It represents the average of the maximum outdoor temperature in month m in the N-year history. It represents the average of the minimum outdoor temperature in month m in the N-year history. represents the standard deviation of the maximum outdoor temperature in month m in historical N years, represents the standard deviation of the minimum outdoor temperature in month m in historical years N; d represents the total number of days in month m; 1.4) The standard coefficients of the highest and lowest outdoor temperatures in each month of the history are calculated through standardization, and the weighted sum is used to obtain the stability coefficient of each month of the year: in, The standard coefficient representing the maximum outdoor temperature in the mth month of the yth year, The standard coefficient representing the minimum outdoor temperature in the mth month of the yth year, represents the stability coefficient of the mth month of the yth year, and the weights of the maximum outdoor temperature and the minimum outdoor temperature are both 0.5; 1.5) For each month m, select the stability coefficient of the mth month of each year in the N historical years The daily outdoor temperature of the whole month with the minimum value is taken as the outdoor temperature data of the mth month of the typical meteorological year; 1.6) Calculate the daily cooling degree days CDD and daily heating degree days HDD according to the following formula: in, is the maximum temperature on the jth day in a typical meteorological year, is the minimum temperature on the jth day in a typical meteorological year, is the preset threshold; S1.2.

2. Use the date type, season type, historical daily carbon emissions of sub-items, daily cooling degree days CDD, and daily heating degree days HDD obtained in step S1.2.1 to train the prediction algorithm, where the feature values ​​can be selected based on experience, such as daily cooling degree days CDD and daily heating degree days HDD, date type, and season type, and the learning input sample is the historical daily carbon emissions data of sub-items collected in step S1.1; the trained prediction algorithm can be used to predict the daily carbon emissions data of the corresponding sub-items; S1.

3. Calculation of climate-independent monthly forecast carbon emissions: S1.3.

1. For the sub-items with low correlation with outdoor temperature identified in step S1.1, for the month m to be predicted, directly use the arithmetic mean and average change rate of the monthly carbon emissions of the corresponding months in all historical years to calculate the monthly predicted carbon emissions, and then proceed to step S1.3.2 to split the sub-item daily carbon emissions; the calculation of the monthly predicted carbon emissions is shown in the following formula: in, represents the monthly carbon emissions of the ith item in the mth month of the yth year in history, represents the average monthly carbon emissions of the ith item in the mth month of the Nth year in history, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the rate of change of the ith item in the mth month of the yth year, It represents the average change rate of the ith sub-item in all years in the mth month; S1.3.

2. Breakdown of daily carbon emissions by climate-independent sub-item: For the monthly forecast carbon emissions calculated in step S1.3.1, the daily proportion coefficient is calculated using the sliding week coefficient method, and then the monthly forecast carbon emissions are split into daily forecast carbon emissions according to the daily proportion coefficient; only 7 days forward every day is a sliding week, and each sliding week includes the completed Monday to Sunday; the calculation process is shown in the following formula: ] Among them, i represents the i-th item, j represents the j-th day of each month, and w represents the w-th day among the 1st to 7th days in the sliding week with the j-th day as the first day. Indicates the sliding weekday ratio coefficient, Indicates the actual daily carbon emissions of the sub-item on the first day of the sliding week. Indicates the actual carbon emissions of each day in the historical sliding week. represents the daily ratio coefficient for the jth day of each month, represents the monthly predicted carbon emissions of the ith item in the mth month, It represents the sum of the daily proportion coefficients of the ith sub-item of the month. It represents the daily predicted carbon emissions of the jth day of the ith sub-item of the month; S2. Issue early warning and / or alarm based on the obtained daily carbon emissions by item.

2. The method according to claim 1, characterized in that In step S1.1, the calculation of the correlation coefficient between the daily carbon emissions of each building item and the outdoor temperature specifically includes: in, Historical daily carbon emissions and historical outdoor temperature The correlation coefficient of , They represent historical daily carbon emissions and historical outdoor temperature data respectively. High correlation includes positive correlation and negative correlation. When the correlation coefficient is in the range of [0.4,1], it means that the daily carbon emissions of the corresponding sub-item are positively correlated with the outdoor temperature; when the correlation coefficient is in the range of [-1,-0.4], it means that the daily carbon emissions of the corresponding sub-item are negatively correlated with the outdoor temperature; when the correlation coefficient is in the range of (-0.4,0.4), it means that the correlation between the daily carbon emissions of the corresponding sub-item and the outdoor temperature is low.

3. The method according to claim 1, characterized in that In step 1.6), , =18 .

4. The method according to claim 3, characterized in that In step S1.3.2, if there is no saved sliding week sub-item daily proportion coefficient, you need to wait for a week before using it, and the 7 sub-item daily proportion coefficients of the first 7-day cycle containing a complete Monday to Sunday will be used as the basis; if there are more than 1 week of data saved, the most recent sliding week will be used as the basis; if there are bad numbers, the most recent sliding week data with complete good numbers will be used to recalculate the sliding week's daily proportion coefficient, and then calculate the sub-item daily predicted carbon emissions.

5. The method according to claim 4, characterized in that In step S2, the specific process of early warning is as follows: 2.1.

1. Calculate the risk rate: Calculate the simulated monthly carbon emissions by item as follows : Among them, h represents the hth day, Indicates the sum of the actual daily carbon emissions of the sub-items in the past days. It represents the daily predicted carbon emissions of the item on that day. Indicates the actual carbon emissions of yesterday. It represents the daily forecast carbon emissions of yesterday. express ; Calculate the carbon emissions progress value as follows: Indicates the monthly itemized carbon emissions budget; The risk rate is calculated as follows: Risk Rate Among them, when When ; 2.1.

2. According to the carbon emission progress calculated in step 2.1.1, the sensitivity is judged and the risk level is judged according to the risk rate.

6. The method according to claim 5, characterized in that In step S2, the specific process of alarming is as follows: 2.2.

1. When the actual carbon emissions per item per month is greater than the budgeted carbon emissions per item per month, an alarm is generated; 2.2.

2. Determine the severity of the alarm: The excess rate is calculated as follows: Exceeding standard rate in, is the actual carbon emissions per month, Itemize carbon emissions budget for the month; The sensitivity of the alarm and the severity of the alarm are determined based on the excess rate.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

8. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program; when the processor is used to execute the computer program, the method described in any one of claims 1 to 6 is implemented.

Images