Total load analysis method, device and application in the design phase of residential quarters using interval prediction

Abstract

The present invention discloses a total load analysis method, device, and application for residential district design using interval prediction. The method involves obtaining historical data from several residential districts within a set range near a new residential district and using this data as calculation data and input training sample data. Time-series load external characteristic power data is obtained based on the calculation data. Maximum load data and time-series equivalent load data are determined from the time-series load external characteristic power data and used as output training sample data. A multi-input single-output interval prediction model and a multi-input multi-output interval prediction model are constructed, respectively. The prediction models are trained using input and output training samples. Design data is input into the trained prediction models to obtain maximum load interval data and time-series equivalent load interval data. The upper limits of the two data are compared, and the larger value is taken as the total load of the new residential district. Using historical data to obtain a reasonable calculated load addresses the overly conservative issue of empirical methods.

Description

Technical Field

[0001] This invention relates to the field of load calculation technology in building design, specifically to a method, apparatus, and application for total load analysis in the design phase of residential communities based on interval prediction. Background Art

[0002] To save energy and reduce carbon emissions, the penetration rate of distributed energy on the power grid user side is constantly increasing, and new loads such as electric vehicles are gradually increasing. However, in the design schemes of existing residential communities, the installed capacity of residential communities is based on empirical values, which are often too large relative to the actual demand, and the impact of emerging factors such as building-integrated photovoltaics and electric vehicles is not taken into account.

[0003] Currently, existing load calculation methods generally employ empirical methods such as the unit index method, demand factor method, and utilization factor method. The unit index method relies on data summarization and is simple to calculate, thus having low accuracy and being only suitable for estimation. The demand factor method and utilization factor method mainly rely on empirical coefficients for calculation, requiring knowledge of the power of all equipment. However, the equipment in the design stage and the actual use stage are not completely consistent, resulting in large errors in some scenarios.

[0004] Forecasting methods offer a solution to the aforementioned problems. These methods utilize techniques such as statistics, machine learning, and data mining to model historical data and predict future trends of events. Point forecasting methods include regression forecasting, classification forecasting, and time series forecasting, while interval forecasting, building upon point forecasting, uses probability theory to derive a forecast interval at a certain confidence level. To ensure sufficient backup equipment capacity, using the upper limit of the interval forecast as a conservative total load calculation value is effective. Therefore, designing a new method for calculating the total load of residential communities to provide a reference for community equipment configuration has become an urgent problem to be solved. Summary of the Invention

[0005] To address this, the present invention provides a method, apparatus, and application for total load analysis in the design phase of residential communities using interval forecasting. This method utilizes historical data to obtain a reasonable calculated load based on interval forecasting, thus solving the problem of overly conservative empirical methods.

[0006] To achieve the above objectives, the present invention provides the following technical solution: S1. Obtain historical data of several residential communities within a set range near the new residential community according to a set data source; use the obtained historical data as calculation data and input training sample data;

[0007] S2. Based on the calculated data, obtain the time-series load external characteristic power data of several nearby residential communities through calculation; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; use the maximum load data and the time-series equivalent load data as output training sample data;

[0008] S3. Construct a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data;

[0009] S4. Input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions, and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community.

[0010] S5. Compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

[0011] As a preferred embodiment of the total load analysis method for the design phase of residential communities based on interval prediction, in the process of using the acquired historical data as the calculation data and the input training sample data, the calculation data includes: time series power data of several nearby residential communities, time series power data of building photovoltaics, and time series power data of electric vehicle charging and discharging.

[0012] The input training sample data includes: the number of households, building area, roof area, and number of charging piles in several nearby residential communities.

[0013] As a preferred embodiment of the total load analysis method for the design phase of residential communities based on interval prediction, in the process of constructing the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model respectively, the maximum load data of several nearby residential communities is used as the output sample data to construct the multi-input single-output interval prediction model; and the time-series equivalent load interval data of several nearby residential communities is used as the output sample data to construct the multi-input multi-output interval prediction model.

[0014] As a preferred option for the total load analysis method in the design phase of residential communities based on interval prediction, in the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the design data of the new residential community includes: the number of households, building area, roof area and number of charging piles of the new residential community.

[0015] As a preferred method for total load analysis in the design phase of residential communities based on interval prediction, a confidence level of 95% is set when the design data of the new residential community is input into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction.

[0016] This invention also provides a total load analysis device for the design phase of residential communities based on interval prediction, which, based on the above-mentioned total load analysis method for the design phase of residential communities based on interval prediction, includes:

[0017] The computational data and input training sample data acquisition module is used to acquire historical data of several residential communities within a set range near the new residential community based on a set data source; and to use the acquired historical data as computational data and input training sample data.

[0018] The output training sample data acquisition module is used to obtain the time-series load external characteristic power data of several nearby residential communities by calculation based on the calculated data; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; and use the maximum load data and the time-series equivalent load data as output training sample data.

[0019] The prediction model building module is used to build a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; and to train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data.

[0020] The prediction interval data acquisition module is used to input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community.

[0021] The total load acquisition module is used to compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

[0022] As a preferred embodiment of the total load analysis device for the design phase of residential communities based on interval prediction, in the calculation data and input training sample data acquisition module, when the acquired historical data is used as the calculation data and the input training sample data, the calculation data includes: time-series power data of several nearby residential communities, time-series power data of building photovoltaics, and time-series power data of electric vehicle charging and discharging.

[0023] The input training sample data includes: the number of households, building area, roof area, and number of charging piles in several nearby residential communities.

[0024] As a preferred embodiment of the total load analysis device for the design phase of a residential community based on interval prediction, in the prediction model construction module, during the construction of the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model, the maximum load data of several nearby residential communities is used as output sample data to construct the multi-input single-output interval prediction model; and the time-series equivalent load interval data of several nearby residential communities is used as output sample data to construct the multi-input multi-output interval prediction model.

[0025] As a preferred embodiment of the total load analysis device for the design phase of a residential community based on interval prediction, in the prediction interval data acquisition module, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the design data of the new residential community includes: the number of households in the new residential community, the building area data, the building roof area data, and the number of charging piles.

[0026] As a preferred embodiment of the total load analysis device for the design phase of residential communities based on interval prediction, in the prediction interval data acquisition module, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the confidence level is set to 95%.

[0027] A computer-readable storage medium includes instructions that, when executed on a computer, cause the computer to perform the above-described method for total load analysis in the design phase of a residential community using interval forecasting.

[0028] This invention has the following advantages: Historical data of several residential communities within a set range near a new residential community are obtained based on a set data source; the obtained historical data is used as calculation data and input training sample data; based on the calculation data, time-series load external characteristic power data of the nearby residential communities are obtained through calculation; the maximum load data and time-series equivalent load data of the nearby residential communities are determined through the time-series load external characteristic power data; the maximum load data and the time-series equivalent load data are used as output training sample data; a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model are constructed respectively; the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model are trained using the input training sample data and the output training sample data; the design data of the new residential community is input into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction to obtain the maximum load interval data and time-series equivalent load interval data of the new residential community; the upper limit value of the maximum load interval data of the new residential community is compared with the upper limit value of the time-series equivalent load interval data, and the larger of the two is taken as the total load of the new residential community. The present invention provides a method for calculating the total load during the design phase of a residential community. After acquiring historical data from other communities within a defined range near the new residential community, the method processes the data and constructs and trains multi-input single-output and multi-input multi-output interval prediction models. Then, based on the design data of the new residential community, it predicts two types of interval data. Finally, the maximum value of the upper bound is compared and taken as the total load. Considering that existing methods for calculating the total load during the design phase rely on data induction and empirical coefficients, their accuracy is low, errors are large in certain scenarios, and the results are overly conservative, leading to unnecessary waste. Therefore, the present invention, based on interval prediction, solves these problems. It models historical data from other residential communities and constructs two interval prediction models with the maximum load and time-series equivalent load as outputs. The prediction yields two sets of relatively conservative interval data, and to ensure a certain reserve capacity, the maximum value of the upper bound of these two models is used as the final calculation result. This ensures both accuracy and redundancy in actual production, while eliminating the need for tedious manual calculations and experience-based coefficient selection. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the total load analysis method for the design stage of a residential community using interval prediction provided in Embodiment 1 of the present invention;

[0030] Figure 2 This is a schematic diagram illustrating the specific steps of the total load analysis method for the design phase of a residential community using interval prediction provided in Embodiment 1 of the present invention.

[0031] Figure 3This is a schematic diagram of the interval prediction results and actual cell requirements in one possible embodiment of Embodiment 1 of the present invention;

[0032] Figure 4 This is a schematic diagram of the total load analysis device architecture for the design phase of a residential community based on interval prediction, provided in Embodiment 2 of the present invention. Detailed Implementation

[0033] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Example 1

[0035] See Figure 1 and Figure 2 Embodiment 1 of the present invention provides a method for total load analysis in the design phase of residential communities based on interval prediction, including the following steps:

[0036] S1. Obtain historical data of several residential communities within a set range near the new residential community based on a set data source; use the obtained historical data as calculation data and input training sample data;

[0037] S2. Based on the calculated data, obtain the time-series load external characteristic power data of several nearby residential communities through calculation; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; use the maximum load data and the time-series equivalent load data as output training sample data;

[0038] S3. Construct a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data;

[0039] S4. Input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions, and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community.

[0040] S5. Compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

[0041] In this embodiment, in step S1, during the process of using the acquired historical data as the calculation data and the input training sample data, the calculation data includes: time-series power data of several nearby residential communities, time-series power data of building photovoltaics, and time-series power data of electric vehicle charging and discharging.

[0042] The input training sample data includes: the number of households, building area, roof area, and number of charging piles in several nearby residential communities.

[0043] Specifically, based on the set data source, historical data from multiple other residential communities near the newly built community is obtained, including resident household data n. house Building area data S building Building roof area data S roof Number of charging piles n charge Residential time-series power data P building (t), Building-in-the-Sky Photovoltaic Time-Series Power Data P PV (t) and electric vehicle charging and discharging time-series power data P EV (t).

[0044] The input training samples are set to the number of households in other residential communities, n. house Building area data S building Building roof area data S roof and the number of charging piles n charge The above data is relatively easy to obtain during the design phase, so it is used as the input training sample.

[0045] The calculation data is set to the time-series power data P of residents in other residential communities. building (t), Building-in-the-Sky Photovoltaic Time-Series Power Data P PV (t) and electric vehicle charging and discharging time-series power data P EV (t), the above data is strongly correlated with the actual load size of the cell in the actual scenario, so these data are used as the calculation data.

[0046] In this embodiment, in step S2, based on the calculated data, the time-series load external characteristic power data P of several nearby residential communities is obtained by calculation. LC (t);

[0047] Specifically, P is based on calculation data from other residential communities. building (t), P PV (t) and P EV The time-series load external characteristic power data P is obtained by calculating (t) using the following formula. LC (t):

[0048] P LC(t)=P building (t)-P PV (t)+P EV (t)

[0049] Time-series load external characteristic power data P LC (t) represents the magnitude of the characteristic load outside the community in the actual scenario.

[0050] The maximum load data and time-series equivalent load data of several nearby residential communities are determined by the aforementioned time-series load external characteristic power data.

[0051] Specifically, based on P LC (t) Further filtering to determine the time-series equivalent load data P of other residential communities. LE (t), for P building (t) is used to filter and determine the maximum load data P of other residential communities. L,max P L,max For single-point data, and P LE (t) represents time-series data; considering both simultaneously can fully encompass various characteristics of the forecast; maximum load data P L,max The calculation formula is:

[0052] P L,max =max(P building (t))

[0053] Both are used as output training samples for two different prediction models, and both single-point and temporal features are included.

[0054] In this embodiment, in step S3, during the construction of the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model, the maximum load data of several nearby residential communities is used as output sample data to construct the multi-input single-output interval prediction model; the time-series equivalent load interval data of several nearby residential communities is used as output sample data to construct the multi-input multi-output interval prediction model.

[0055] Specifically, construct a P L,max A multi-input single-output interval prediction model for output samples is used to represent single-point features; a model is constructed with P... LE (t) represents the multi-input multi-output interval prediction model for the output samples, reflecting the temporal characteristics;

[0056] The multi-input single-output interval prediction model and the multi-input multi-output interval prediction model are trained using the input training sample data and the output training sample data.

[0057] In this embodiment, in step S4, the design data of the new residential community is input into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, so as to obtain the maximum load interval data and time-series equivalent load interval data of the new residential community.

[0058] Specifically, obtain the design data for the new residential community, including the number of households in the new residential community, n. house,new Building area data S building,new Building roof area data S roof,new and the number of charging piles n charge,new ;

[0059] The design data of the new residential community is input into two prediction models to perform interval prediction with multiple inputs and single outputs and multiple inputs and multiple outputs.

[0060] To ensure sufficient reserve capacity while avoiding overly conservative results, a 95% confidence level was used for interval forecasting, yielding two forecast results. These are the predicted maximum load interval data for the new residential community: R = (P L,max,lower ,P L,max,upper ) and the predicted time-series equivalent load interval data S=(P LE,lower (t),P LE,upper (t)).

[0061] In this embodiment, in step S5, the upper limit of the maximum load interval data of the new residential community is compared with the upper limit of the time-series equivalent load interval data, and the larger of the two is taken as the total load of the new residential community.

[0062] Specifically, to ensure a certain reserve capacity, the maximum value P of the upper limit of the data range for the time-series equivalent load is selected. LE,upper (t) max As a selection of prediction results, the upper limit value P of the maximum load data interval is selected. L,max,upper As an alternative prediction outcome; for P LE,upper (t) max and P L,max,upper The two values ​​are compared, and the maximum value is taken as the total load P. Load .

[0063] The interval forecasting method selected in this invention is the Least Squares Support Vector Machine (LS-SVM) method, which is a variant of Support Vector Machine (SVM) and is currently used to solve various problems such as load forecasting and fault classification. Therefore, this invention chooses this method for description, but various other interval forecasting methods are also applicable to this invention, and no specific limitation is made thereto.

[0064] In one possible embodiment, the following calculation example is provided based on a newly built cell and its nearby cells:

[0065] After obtaining historical data from multiple communities within a defined range near the new residential community, the data is divided into computational data and input training samples. The computational data is processed to obtain the time-series load external characteristic power of the multiple nearby communities. The maximum load data and time-series equivalent load data are determined as output training samples. Based on the LSSVM interval prediction method and the obtained input and output training samples, two prediction models are constructed and trained respectively. Then, data from the design phase of the new residential community is obtained and substituted into the trained model to perform multi-input single-output and multi-input multi-output predictions. The predicted maximum load interval data and the predicted time-series equivalent load interval data of the new residential community are obtained with a 95% confidence level. The maximum value of the upper limit of the time-series equivalent load data interval is compared with the upper limit of the maximum load data interval, and the maximum value is taken as the total load.

[0066] like Figure 3 As shown, the interval prediction results and the actual demand of the community show that the upper limit of the maximum load interval is 4466.819kW, and the maximum value of the upper limit of the time-series equivalent load data interval is 5023.500kW. 5023.500kW is selected as the total load. Compared with the actual demand of 4224.533kW, it can be seen that the total load calculated by this invention is close to the actual power demand while ensuring a certain reserve capacity, thus avoiding a large amount of redundancy in the actual installed capacity.

[0067] In summary, this invention obtains historical data from several residential communities within a defined range near a new residential community based on a set data source; uses the obtained historical data as calculation data and input training sample data; calculates the time-series load external characteristic power data of the nearby residential communities based on the calculation data; determines the maximum load data and time-series equivalent load data of the nearby residential communities based on the time-series load external characteristic power data; uses the maximum load data and the time-series equivalent load data as output training sample data; constructs a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; trains the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data; inputs the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction to obtain the maximum load interval data and time-series equivalent load interval data of the new residential community; compares the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and takes the larger value as the total load of the new residential community. The present invention provides a method for calculating the total load during the design phase of a residential community. After acquiring historical data from other communities within a defined range near the new residential community, the method processes the data and constructs and trains multi-input single-output and multi-input multi-output interval prediction models. Then, based on the design data of the new residential community, it predicts two types of interval data. Finally, the maximum value of the upper bound is compared and taken as the total load. Considering that existing methods for calculating the total load during the design phase rely on data induction and empirical coefficients, their accuracy is low, errors are large in certain scenarios, and the results are overly conservative, leading to unnecessary waste. Therefore, the present invention, based on interval prediction, solves these problems. It models historical data from other residential communities and constructs two interval prediction models with the maximum load and time-series equivalent load as outputs. The prediction yields two sets of relatively conservative interval data, and to ensure a certain reserve capacity, the maximum value of the upper bound of these two models is used as the final calculation result. This ensures both accuracy and redundancy in actual production, while eliminating the need for tedious manual calculations and experience-based coefficient selection.

[0068] It should be noted that the method of this disclosure embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this disclosure embodiment, and the multiple devices will interact with each other to complete the method described.

[0069] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] Example 2

[0071] See Figure 4 Embodiment 2 of the present invention also provides a total load analysis device for the design stage of residential communities based on interval prediction, comprising:

[0072] The calculation data and input training sample data acquisition module 001 is used to acquire historical data of several residential communities within a set range near the new residential community according to a set data source; and to use the acquired historical data as calculation data and input training sample data.

[0073] The output training sample data acquisition module 002 is used to obtain the time-series load external characteristic power data of several nearby residential communities by calculation based on the calculated data; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; and use the maximum load data and the time-series equivalent load data as output training sample data.

[0074] The prediction model construction module 003 is used to construct a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; and to train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data.

[0075] The prediction interval data acquisition module 004 is used to input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community.

[0076] The total load acquisition module 005 is used to compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

[0077] In this embodiment, in the calculation data and input training sample data acquisition module 001, during the process of using the acquired historical data as the calculation data and the input training sample data, the calculation data includes: time-series power data of several nearby residential communities, building photovoltaic time-series power data, and electric vehicle charging and discharging time-series power data.

[0078] The input training sample data includes: the number of households, building area, roof area, and number of charging piles in several nearby residential communities.

[0079] In this embodiment, in the prediction model construction module 003, during the construction of the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model, the maximum load data of several nearby residential communities is used as output sample data to construct the multi-input single-output interval prediction model; and the time-series equivalent load interval data of several nearby residential communities is used as output sample data to construct the multi-input multi-output interval prediction model.

[0080] In this embodiment, in the prediction interval data acquisition module 004, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the design data of the new residential community includes: the number of households in the new residential community, the building area data, the building roof area data, and the number of charging piles.

[0081] At the same time, the confidence level was set at 95%.

[0082] It should be noted that the information interaction and execution process between the modules of the above system are based on the same concept as the method embodiment in Embodiment 1 of this application, and the resulting technical effects are the same as those in the method embodiment of this application. For details, please refer to the description in the method embodiment shown above in this application, and it will not be repeated here.

[0083] Example 3

[0084] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium storing program code for a total load analysis method for the design phase of a residential community based on interval prediction. The program code includes instructions for executing the total load analysis method for the design phase of a residential community based on interval prediction as described in Embodiment 1 or any possible implementation thereof.

[0085] Computer-readable storage media can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives, SSDs).

[0086] Example 4

[0087] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0088] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor can call the program instructions to execute the total load analysis method for the design phase of residential communities based on interval prediction, as described in Embodiment 1 or any possible implementation thereof.

[0089] Specifically, a processor can be implemented in hardware or software. When implemented in hardware, the processor can be a logic circuit, an integrated circuit, etc. When implemented in software, the processor can be a general-purpose processor that reads software code stored in memory. This memory can be integrated into the processor or located outside the processor and exist independently.

[0090] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0091] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing systems. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Optionally, they can be implemented using program code executable by a computing system, thereby storing them in a storage system for execution by the computing system. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0092] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for total load analysis in the design phase of residential communities using interval forecasting, characterized in that, The following steps are involved: S1. Obtain historical data of several residential communities within a set range near the new residential community based on a set data source; use the obtained historical data as calculation data and input training sample data; S2. Based on the calculated data, obtain the time-series load external characteristic power data of several nearby residential communities through calculation; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; use the maximum load data and the time-series equivalent load data as output training sample data; S3. Construct a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data; S4. Input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions, and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community. S5. Compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

2. The method for total load analysis in the design phase of residential communities using interval forecasting as described in claim 1, characterized in that, In S1, the calculation data includes time-series power data of several nearby residential communities, time-series power data of building photovoltaics, and time-series power data of electric vehicle charging and discharging; the input training sample data includes data on the number of households, building area, roof area, and number of charging piles of several nearby residential communities.

3. The method for total load analysis in the design phase of residential communities using interval forecasting as described in claim 1, characterized in that, In S3, during the construction of the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model, the maximum load data of several nearby residential communities is used as the output sample data to construct the multi-input single-output interval prediction model; the time-series equivalent load interval data of several nearby residential communities is used as the output sample data to construct the multi-input multi-output interval prediction model.

4. The method for total load analysis in the design phase of residential communities using interval forecasting as described in claim 1, characterized in that, In S4, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the design data of the new residential community includes the number of households, building area, roof area and number of charging piles.

5. The method for total load analysis in the design phase of residential communities using interval forecasting as described in claim 1, characterized in that, In S4, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the confidence level is set to 95%.

6. A total load analysis device for the design phase of a residential community using interval forecasting, employing the total load analysis method for the design phase of a residential community using interval forecasting as described in any one of claims 1-5, characterized in that, include: The computational data and input training sample data acquisition module is used to acquire historical data of several residential communities within a set range near the new residential community based on a set data source. The acquired historical data is used as computational data and input training sample data; The output training sample data acquisition module is used to obtain the time-series load external characteristic power data of several nearby residential communities by calculation based on the calculated data; determine the maximum load data and time-series equivalent load data of several nearby residential communities through the time-series load external characteristic power data; and use the maximum load data and the time-series equivalent load data as output training sample data. The prediction model building module is used to build a multi-input single-output interval prediction model and a multi-input multi-output interval prediction model respectively; and to train the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model using the input training sample data and the output training sample data. The prediction interval data acquisition module is used to input the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model to make predictions and obtain the maximum load interval data and time-series equivalent load interval data of the new residential community. The total load acquisition module is used to compare the upper limit of the maximum load interval data of the new residential community with the upper limit of the time-series equivalent load interval data, and take the larger value as the total load of the new residential community.

7. The total load analysis device for the design stage of residential communities using interval prediction as described in claim 6, characterized in that, In the module for acquiring computational data and input training sample data, when the acquired historical data is used as the computational data and the input training sample data, the computational data includes: time-series power data of several nearby residential communities, time-series power data of building photovoltaics, and time-series power data of electric vehicle charging and discharging; the input training sample data includes data on the number of households, building area, roof area, and number of charging piles in several nearby residential communities.

8. The total load analysis device for the design stage of residential communities using interval prediction as described in claim 6, characterized in that, In the prediction model construction module, during the construction of the multi-input single-output interval prediction model and the multi-input multi-output interval prediction model, the maximum load data of several nearby residential communities is used as output sample data to construct the multi-input single-output interval prediction model; the time-series equivalent load interval data of several nearby residential communities is used as output sample data to construct the multi-input multi-output interval prediction model. In the prediction interval data acquisition module, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the design data of the new residential community includes: the number of households, building area, roof area, and number of charging piles in the new residential community. In the prediction interval data acquisition module, during the process of inputting the design data of the new residential community into the trained multi-input single-output interval prediction model and the multi-input multi-output interval prediction model for prediction, the confidence level is set to 95%.

9. An information data processing terminal that implements the total load analysis method for the design phase of a residential community using interval prediction as described in any one of claims 1-5.

10. A computer-readable storage medium comprising instructions, when executed on a computer, causing the computer to perform the total load analysis method for the design phase of a residential community using interval forecasting as described in any one of claims 1-5.

Images