A method and device for predicting the flexible regulation potential of air conditioning in residential buildings based on uncertainty

Abstract

The present invention discloses a method and device for predicting the flexible regulation potential of air conditioning in residential complexes based on uncertainty. The method comprises obtaining information related to flexible energy use of air conditioning in residential complexes through household surveys; predicting the random air conditioning usage time and set temperature, equipment energy efficiency, and flexible regulation willingness of all residents in the residential complex; implementing a flexible regulation strategy for the residential complex's air conditioning based on the aforementioned energy use behavior, equipment performance, and resident willingness, and predicting energy consumption before and after flexible regulation of the residential complex's air conditioning; and predicting the flexible regulation potential of the residential complex's air conditioning based on a flexible regulation potential evaluation index. The method, with its simple modeling rules and high accuracy, is particularly suitable for predicting the flexible regulation potential of air conditioning in residential complexes characterized by highly random air conditioning energy use behavior, diverse equipment performance, and uncertain resident flexible regulation willingness.

Description

Technical Field

[0001] The present invention relates to the fields of big data, regional energy planning, heating, ventilation and air conditioning, and in particular to a method and device for predicting the flexible adjustment potential of air conditioning in residential building complexes based on uncertainty. Background Art

[0002] The traditional "source follows load" stable power grid model is no longer able to cope with the volatility of renewable energy on the supply side. Demand-side energy flexibility is the ability of households to flexibly adjust their energy needs at different times based on the energy network's production capacity and supply conditions. The new "source-load interaction" power grid model, through the efficient scheduling of flexible resources on the demand side, has become an effective way to cope with renewable energy fluctuations and alleviate wind and solar power curtailment.

[0003] Residential buildings are a crucial component of building energy conservation and emission reduction. By integrating a certain number of households within a region to participate in electricity market demand response, load aggregators can significantly increase the amount of flexible load regulation in residential buildings, demonstrating enormous potential for regulation. However, the numerous uncertainties and high randomness of flexible energy use in residential complexes make accurate prediction of their energy flexibility potential difficult, creating a significant bottleneck restricting flexible energy use in urban residences. Air conditioning equipment is a key flexible energy consumer in residential buildings, accounting for approximately 40% of total energy consumption. It can achieve flexible regulation by adjusting set temperatures, operating hours, and utilizing the thermal inertia of the building envelope to store heat and cool air, significantly enhancing energy flexibility. However, most research on air conditioning equipment focuses on the flexible energy use of centralized air conditioning systems in public buildings, and research on the flexible energy use of air conditioning equipment commonly used in residential buildings is scarce.

[0004] The few existing studies on the dynamic prediction of residential air conditioning energy flexibility potential primarily predict the dynamic load, energy consumption, and system performance changes before and after the implementation of a regulation strategy, and then calculate the flexible regulation potential of air conditioning equipment based on flexible regulation potential evaluation indicators. Currently, only a few researchers have examined the impact of the randomness of air conditioning usage time on residential energy consumption before the implementation of the regulation strategy (i.e., reference operating conditions). However, no research has yet been conducted on the impact of the randomness of air conditioning set temperatures, the diversity of air conditioning equipment energy efficiency, and the uncertainty of residents' flexible regulation willingness on the prediction of residential air conditioning flexibility potential. Consequently, these studies fail to reflect the impact of various uncertainties on flexible regulation potential, making it difficult to accurately predict the flexible regulation potential of residential air conditioning intensities. Summary of the Invention

[0005] In view of the fact that existing methods cannot reflect the uncertainty of air-conditioning usage time, set temperature, equipment energy efficiency and residents' willingness to adjust air conditioners, thus affecting the accurate prediction of the flexible adjustment potential of air conditioners in residential buildings, the present invention proposes a method and device for predicting the flexible adjustment potential of air conditioners in residential buildings based on uncertainty. The flexible energy consumption information of air conditioners in residential buildings is obtained through household survey method; a random prediction model for air conditioner usage time in residential buildings is constructed by K-Means clustering and Monte Carlo method; a random prediction model for air conditioner set temperature in residential buildings and a random prediction model for energy efficiency of residential equipment are constructed by Monte Carlo method; a baseline energy consumption prediction model for air conditioners in residential buildings is constructed by EnergyPlus energy consumption simulation software; a random prediction model for residents' willingness to adjust air conditioners is constructed by Monte Carlo method; based on residents' willingness, a flexible adjustment strategy decision for air conditioners in residential buildings is made, and an energy consumption prediction model for air conditioners in residential buildings after flexible adjustment is constructed by EnergyPlus energy consumption simulation software; combined with flexible adjustment potential evaluation index, the flexible adjustment potential of air conditioners in residential buildings is calculated based on the predicted energy consumption of air conditioners before and after the implementation of the flexible adjustment strategy.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows: a method for predicting the flexible adjustment potential of air conditioning in residential buildings based on uncertainty, wherein the specific steps of the prediction method are as follows:

[0007] Step S1: Acquire flexible energy consumption information of air conditioners in residential buildings; obtain the air conditioner usage time and set temperature, air conditioner equipment performance, indoor and outdoor temperatures, and residents' flexible adjustment willingness to increase the cooling set temperature or decrease the heating set temperature for different residents in the residential area; extract typical characteristics of air conditioner usage time and calculate the distribution probability of various typical air conditioner usage time patterns; calculate the distribution probability of air conditioner set temperature and equipment energy efficiency; calculate the distribution probability of residents' flexible adjustment willingness; and randomly predict the air conditioner usage time and set temperature, equipment energy efficiency, and flexible adjustment willingness of all residents in the residential area using the Monte Carlo method.

[0008] Step S2: Construct a prediction model for the baseline energy consumption of air conditioning, cooling, and heating for the residential complex; based on the randomly predicted air conditioning usage time, set temperature, and equipment energy efficiency, complete the prediction of the baseline energy consumption of air conditioning, cooling, and heating for the residential complex through energy consumption simulation in the cooling and heating seasons;

[0009] Step S3: Based on the randomly predicted flexible adjustment intentions of all residents in the residential area, a flexible adjustment strategy is determined for each air conditioner in each household during the cooling and heating seasons, and the set temperature of the air conditioner after flexible adjustment is obtained;

[0010] Step S4: Constructing an energy consumption prediction model for the air conditioners in the residential complex after flexible adjustment; based on the set temperature after flexible adjustment obtained in step S3, the cooling and heating energy consumption predictions for the air conditioners in the residential complex after flexible adjustment are completed by simulating the energy consumption in the corresponding cooling and heating seasons;

[0011] Step S5: Determine the flexible regulation potential evaluation index; based on the energy consumption prediction results of steps S2 and S4, predict the flexible regulation potential of the air conditioning of the residential building complex.

[0012] Furthermore, in step S1, a random prediction model for air-conditioning usage behavior in residential buildings is constructed; typical characteristics of air-conditioning usage time in residential buildings are extracted and surveyed to obtain typical air-conditioning usage time patterns in the buildings; based on the probabilities of various typical patterns, the random air-conditioning usage time of all residents in the residential area is predicted separately by the Monte Carlo method.

[0013] Furthermore, in step S1, a random prediction model for the air-conditioning set temperature of the residential building complex is constructed; based on the air-conditioning set temperature data collected from the survey, the distribution probability of the air-conditioning set temperature of the residential building complex is calculated; and then the corresponding set temperature is randomly assigned to each air-conditioning of each household through the Monte Carlo method.

[0014] Furthermore, in step S1, a random prediction model for the energy efficiency of air-conditioning equipment in residential buildings is constructed; based on the energy efficiency data of air-conditioning equipment collected from the survey, the distribution probability of various energy efficiency levels of air-conditioning in residential buildings is calculated; and then the corresponding energy efficiency level is randomly assigned to each air-conditioning of each household through the Monte Carlo method.

[0015] Furthermore, in step S1, a random prediction model for flexible adjustment willingness of residents in the residential complex is constructed. Residents' flexible adjustment willingness includes whether the air conditioner participates in flexible adjustment on that day and the flexible adjustment range of the air conditioner set temperature. The flexible adjustment range of the air conditioner set temperature is 1°C to 4°C higher or lowered for heating, respectively, based on the baseline set temperature. Based on the statistics of residents' flexible adjustment willingness collected from the survey, the probability distribution of each type of flexible adjustment willingness is calculated. Then, a Monte Carlo method is used to randomly assign a corresponding flexible adjustment willingness to each resident and each air conditioner to quantify the uncertainty of residents' flexible adjustment willingness.

[0016] Furthermore, in step S2, the architectural layout and thermal performance of the building envelope of the selected residential area are collected, and an energy consumption simulation model of the residential building complex is constructed using the EnergyPlus software; the air-conditioning usage time, air-conditioning set temperature and equipment energy efficiency randomly predicted in step S1 are used as input parameters of the model, and the predicted benchmark energy consumption of air-conditioning of the residential buildings is used as the output parameter of the model.

[0017] Furthermore, in step S3, based on the resident's flexible adjustment willingness randomly predicted in step S1, a flexible adjustment strategy for the air conditioner on that day is determined. Specifically, the flexible adjustment strategy is determined based on the resident's willingness to participate in flexible adjustment on that day. If the flexible adjustment strategy is executed, the set temperature of the air conditioner after flexible adjustment is determined based on the willingness to select the flexible adjustment range of the air conditioner set temperature.

[0018] Furthermore, in step S4, the input parameter is the set temperature of each air conditioner of each household in the residential building after flexible adjustment, which is predicted in step S3; and the output parameter is the energy consumption of the air conditioner in the residential building after flexible adjustment.

[0019] Furthermore, in step S5, the load flexibility, energy flexibility and flexible benefits of the air-conditioning in the residential building complex after the flexible adjustment strategy is implemented are calculated to predict the flexible adjustment potential of the air-conditioning in the residential building complex; load flexibility refers to the change in operating load during the peak electricity price period before and after flexible adjustment; energy flexibility refers to the change in operating energy consumption during the peak electricity price period before and after flexible adjustment; flexible benefit refers to the change in operating cost during the adjustment period before and after flexible adjustment.

[0020] On the other hand, the present invention also provides a device for predicting the potential for flexible adjustment of air conditioning in residential building complexes based on uncertainty, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it implements the method for predicting the potential for flexible adjustment of air conditioning in residential building complexes based on uncertainty.

[0021] The beneficial effects of the present invention are as follows: the modeling rules of the present invention are simple and the accuracy is high, and it is particularly suitable for predicting the flexible adjustment potential of air conditioning in residential buildings taking into account the strong randomness of air conditioning energy consumption behavior, the diversity of equipment performance, and the uncertainty of residents' flexible adjustment willingness. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the uncertainty-based method for predicting the flexible regulation potential of air conditioning in residential complexes provided by the present invention.

[0023] Figure 2 This is a typical air conditioning usage time pattern diagram for a residential building complex in an embodiment of the present invention.

[0024] Figure 3 1 is a probability distribution diagram of the air-conditioning set temperature of the residential building complex in an embodiment of the present invention.

[0025] Figure 4 This is a probability distribution diagram of energy efficiency of air-conditioning equipment in a residential building complex in an embodiment of the present invention.

[0026] Figure 5 It is a probability distribution diagram of the flexible adjustment willingness of residents in the residential area in an embodiment of the present invention.

[0027] Figure 6 This is a prediction effect diagram of the total heating energy consumption of a residential building complex during daily peak hours in an embodiment of the present invention.

[0028] Figure 7 This is a prediction effect diagram of the total cooling energy consumption of a residential building complex during daily peak hours in an embodiment of the present invention.

[0029] Figure 8 This is a prediction effect diagram of the flexibility of cooling and heating loads of residential buildings in an embodiment of the present invention.

[0030] Figure 9 This is a prediction effect diagram of the total air-conditioning energy flexibility of the residential building complex in the cooling season / heating season in an embodiment of the present invention.

[0031] Figure 10 This is a diagram showing the predicted effect of the total air conditioning flexibility benefits of the residential building complex during the cooling season / heating season in an embodiment of the present invention.

[0032] Figure 11 It is a structural schematic diagram of an uncertainty-based residential building complex air-conditioning flexible adjustment potential prediction device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] The present invention provides a method for predicting the flexible regulation potential of air conditioning in residential buildings based on uncertainty, and the specific steps are as follows: Figure 1 Shown, including:

[0035] Step S1: First, select representative residential areas based on factors such as size, construction age, and household economic status. Then, conduct household surveys to obtain flexible energy consumption information related to air conditioning equipment energy efficiency, usage time, set temperatures, and residents' willingness to adjust air conditioning settings. The sample size for the survey was determined using the Cochran formula, as shown below.

[0036]

[0037] Where n is the sample size; E is the margin of error; p is the proportion of the population with the attribute of interest; and z is the standardized score. Commonly used confidence intervals are 90%, 95%, and 99%, corresponding to z-standardized scores of 1.645, 1.96, and 2.575.

[0038] For usage time and set temperature information, surveys can be conducted through either in-home measurements or questionnaires. If in-home measurements are used, the air conditioner's operating power or energy consumption, as well as the set temperature, must be measured. Measurement equipment includes power or electricity meters and temperature meters. Based on the measured power fluctuations, the air conditioner's usage can be inferred to determine usage time. If a questionnaire survey is used, the content of the questionnaire should include daily air conditioner usage time and set temperature.

[0039] To gather information on equipment energy efficiency and residents' willingness to use flexible air conditioning, a questionnaire survey can be conducted. To collect information on equipment energy efficiency, residents are asked to take and upload photos of the energy efficiency rating labels on their air conditioners. To gather information on residents' willingness to use flexible air conditioning, the questionnaire includes information on their willingness to use flexible air conditioning and the range of flexible temperature settings they choose. The survey categorizes residents' willingness to use flexible air conditioning into five levels: "100% willing," "75% willing," "50% willing," "25% willing," and "unwilling." "100% willing" indicates residents are willing to use flexible air conditioning every day of the week, "75% willing" indicates residents are willing to use flexible air conditioning five out of seven days, "unwilling" two days, and so on. Furthermore, residents who are willing to use flexible air conditioning will be asked how much they are willing to adjust their air conditioner set temperatures, such as by 1, 2, 3, or 4 degrees Celsius during peak electricity prices in the summer or by 1, 2, 3, or 4 degrees Celsius during winter.

[0040] Secondly, the K-means clustering algorithm (K-Means) is used to extract the typical characteristics of air conditioner usage time. The K-Means algorithm is often used to classify data sets. Its core idea is to find k cluster centers c1, c2, ..., c k , so that every data x i Point and its nearest cluster center c v The sum of the squared distances is minimized. One of the key steps in using the K-Means clustering method is to determine the optimal number of clusters. The present invention uses the sum of the squared errors (SSE, i.e., the inflection point method) and the Davies-Bouldin index (DBI) to jointly predict the clustering effect under different cluster numbers:

[0041] SSE is the sum of the clustering errors of all samples and represents the quality of the clustering effect. As the number of clusters K increases, the sample division becomes more refined, the degree of aggregation of each cluster gradually increases, and the SSE value decreases sharply, and then the decline tends to be gradual. When the SSE tends to be flat, the corresponding K value is the optimal number of clusters. The relationship between SSE and K can be expressed as follows:

[0042]

[0043] Where K is the number of clusters, Ci is the i-th cluster, q is C i The sample points in m i It is C i The center of mass (i.e. C i The mean of all samples in ).

[0044] DBI is used to represent the average similarity of each cluster. A lower DBI value represents a better clustering result, and a K value corresponding to a smaller DBI value should be selected. The DBI value can be calculated using the following formula:

[0045]

[0046] Where K is the number of clusters, S i is the average distance from each sample in cluster i to its cluster centroid, S j It refers to the average distance between each sample in cluster j and its cluster centroid, M i,j is the distance between the centroid of cluster i and the centroid of cluster j.

[0047] Using the data analysis software IBM SPSS Statistics, the data samples related to the daily hourly operating status of air conditioners obtained from the survey were batch-entered into the software for K-Means clustering. During the analysis process, the number of iterations was set to 100 to ensure that the final clustering results were stable, converged, and no longer changed with increasing iterations. Based on the number of clusters determined by the SSE and DBI indicators, the daily air conditioner usage time was summarized into several typical air conditioner usage time patterns, and the probabilities of each typical pattern were obtained. Each typical air conditioner usage time pattern is characterized by the probability of the air conditioner being turned on at each hour, and each sample has similar usage time distribution characteristics.

[0048] Based on the probabilities of various typical air conditioner usage patterns, the Monte Carlo method was used to randomly generate air conditioner usage patterns for all residents in the case study residential area. Specifically, the Monte Carlo method was used to generate a random number uniformly distributed between [0, 1]. This number was then compared with the probabilities of each typical pattern to determine the typical air conditioner usage pattern for each household. Based on the probability of the air conditioner being on at each hour in each typical pattern, the Monte Carlo method was used to randomly generate a daily air conditioner schedule for each household to describe their daily air conditioner usage behavior. Specifically, the Monte Carlo method was used to generate a random number uniformly distributed between [0, 1] at each hour. This number was then compared with the probability distribution of the air conditioner being on at that hour in the typical pattern. 24 decisions were then made to determine whether the equipment was on for each hour of the day.

[0049] Simultaneously, based on the data collected in step S1, the probability distribution of air conditioner setpoint temperatures in the residential area is calculated. Then, a Monte Carlo method is used to randomly assign a corresponding operating temperature to each device in each household. Specifically, the Monte Carlo method generates a random number uniformly distributed between [0, 1], compares it with the probability distribution of air conditioner setpoint temperatures, and determines the setpoint temperature for that household's air conditioner.

[0050] In addition, based on the data collected in step S1, the probability distribution of energy efficiency ratings for air conditioners in the residential area is calculated. Then, a Monte Carlo method is used to randomly assign a corresponding energy efficiency rating to each device in each household. Specifically, the Monte Carlo method generates a random number uniformly distributed between [0, 1] and compares it with the probability distribution of the energy efficiency ratings of the air conditioners to determine the energy efficiency of the air conditioner for that household.

[0051] Furthermore, based on the data collected from the survey in step S1 regarding residents' willingness to adjust their air conditioning, the distribution probability of their willingness to adjust their air conditioning is calculated. The Monte Carlo method is then used to assign each household's willingness to adjust their air conditioning. Specifically, the Monte Carlo method is used to generate a random number uniformly distributed between [0, 1]. This number is then compared with the probability distribution of the household's willingness to adjust their air conditioning to determine whether the household participates in the adjustment. If the household participates, the Monte Carlo method is used to generate a random number uniformly distributed between [0, 1] on an hourly basis. This number is then compared with the probability distribution of the range of flexible adjustment levels to determine the household's willingness to adjust their air conditioning.

[0052] Step S2: Based on the randomly predicted air conditioning usage time, set temperature and equipment energy efficiency in step S1, input the Energyplus simulation software to simulate the energy consumption in the cooling season and heating season, and complete the prediction of the baseline energy consumption of air conditioning cooling and heating in the residential building complex.

[0053] Step S3: Based on the predicted resident's willingness to participate in flexible adjustment, a daily flexible adjustment strategy is determined for each household's air conditioner during the cooling and heating seasons. Specifically, the flexible adjustment strategy is determined for each air conditioner based on the resident's willingness to participate in flexible adjustment on that day. If the flexible adjustment strategy is implemented, the post-flexible adjustment set temperature is determined based on the resident's willingness to select the flexible adjustment range for the air conditioner's set temperature. The post-flexible adjustment set temperature is calculated by increasing the cooling set temperature or decreasing the heating set temperature by 1°C to 4°C, based on the pre-adjustment set temperature, but within the device's inherent adjustable limit.

[0054] Step S4: The daily flexible adjustment strategy of each air conditioner of each household predicted in step S3 is used as the input of the cooling and heating benchmark energy consumption prediction model of the residential building complex; through the energy consumption simulation of the corresponding cooling season and heating season, the cooling and heating energy consumption prediction of the residential building complex after the air conditioners perform flexible adjustment is completed.

[0055] Step S5: Select three flexible regulation potential evaluation indicators, namely load flexibility, energy flexibility, and flexible benefit, to conduct flexible regulation potential analysis:

[0056] (a) Load flexibility: The change in peak operating load before and after flexible adjustment, ΔP t

[0057] Load flexibility refers to the load change of the air conditioner during the peak period of electricity price on the day. The calculation method is shown in the following formula.

[0058] ΔP t =P t -P t,F

[0059] Among them, P t P is the air conditioning load during the peak period of electricity price before flexible adjustment, in W; t,F is the air conditioning operating load during the peak electricity price period after flexible adjustment, unit: W.

[0060] (b) Energy flexibility: the change in peak operating energy consumption before and after flexible adjustment, ΔE t

[0061] Energy flexibility refers to the range of changes in electricity consumption in time period t and demand k under scenario, and is calculated as shown in the following formula.

[0062] ΔE t =E t -E t,F

[0063] Among them, E t The power consumption of air conditioner operation during peak period t before flexible adjustment, unit: kWh; E t,F The power consumption of air conditioner operation during peak period t after flexible adjustment of electricity price, unit is kWh.

[0064] (c) Flexible benefit: the change in operating cost during the adjustment period before and after flexible adjustment, ΔYt R

[0065] Flexible income refers to the flexible income generated during the flexible adjustment period of air conditioning (covering peak adjustment period and off-peak adjustment period), and the calculation method is shown in the following formula.

[0066] ΔYt R =ΔEt R ×(YY F )

[0067] Among them, Y is the dynamic electricity price of air conditioner operation before flexible adjustment, RMB; F is the dynamic electricity price of air conditioner operation after flexible adjustment, RMB; ΔEt Ris the flexible adjustment period t R The electricity consumption of air conditioner operation, i.e. the sum of the increase in energy consumption in the hour before the peak period compared with the energy consumption before adjustment and the reduction in energy consumption in the peak period compared with the energy consumption before adjustment, in kWh.

[0068] The benchmark energy consumption for cooling and heating of the residential complex predicted in step S2 and the energy consumption for cooling and heating of the residential complex after flexible adjustment of the air conditioner of the residential complex predicted in step S4 are used as input variables of the selected flexible adjustment potential evaluation index. The load flexibility, energy flexibility and flexible benefit after the air conditioner of the residential complex implements the flexible adjustment strategy are calculated to predict the flexible adjustment potential of the air conditioner of the residential complex.

[0069] Based on K-Means clustering, Monte Carlo randomization, and related simulation modeling methods, this paper establishes a comprehensive method for predicting the flexible regulation potential of air conditioning in residential complexes. This method accurately characterizes the uncertainties of air conditioning usage time, set temperature, equipment energy efficiency, and residents' willingness to adjust air conditioning in residential complexes, and incorporates these uncertainties into the prediction of flexible regulation potential for residential complexes. This method is of great significance for demand-side peak shaving, renewable energy consumption, and operational cost savings.

[0070] Example: The present invention takes a residential area in a hot summer and cold winter region as an example, conducts surveys in the residential area from June to September in summer and from December to February in winter, and then uses this method to predict the flexible adjustment potential of air conditioning in the residential area.

[0071] To verify the accuracy of the method, this embodiment uses both on-site measurements and questionnaire surveys to obtain information on flexible energy consumption of air conditioners. First, this embodiment measured the operating power and operating temperature of 122 bedroom air conditioners and 111 living room air conditioners to obtain information on air conditioner usage time and air conditioner set temperature. At the same time, questionnaire surveys on air conditioner usage time and air conditioner set temperature information were distributed to the measured residents. By comparing the results of daily measured usage time with the results of daily questionnaire usage time, the random prediction model of air conditioner usage time in residential buildings was verified. At the same time, by comparing the results of daily measured set temperature with the results of daily questionnaire set temperature, the random prediction model of air conditioner set temperature in residential buildings was verified. For the random prediction model of energy efficiency of air conditioner equipment in residential buildings, no verification is required because both the actual measurement and the questionnaire adopt the method of photographing and uploading energy efficiency marks.

[0072] The present invention uses the coefficient of variation of root meansquare error (CV-RMSE) indicator for calibration, and the calculation method is shown in the following formula.

[0073]

[0074] Among them, ai is the actual target, p i The predicted target value for the simulation. is the average value of the actual measurement value, and n is the number of samples.

[0075] Comparing the measured air conditioning usage time with the questionnaire results, the CV-RMSE for over 95% of households was less than 5%. Comparing the measured air conditioning operating temperature with the questionnaire-set temperature results, the CV-RMSE for the percentage of households using the bedroom air conditioner at 21°C and the living room air conditioner at 30°C was less than 1%, with a slight difference in the percentage of households using the bedroom air conditioner at 21°C and the living room air conditioner at 30°C, respectively, with CV-RMSEs of 4% and 7%, respectively. These results all meet the simulation accuracy test requirement of less than 15% for CV-RMSE as specified in the ASHRAE Measurement of Energy and DemandSavings standard.

[0076] After completing the calibration of the random prediction model for the usage time of air conditioners in residential buildings, the random prediction model for the set temperature of air conditioners in residential buildings, and the random prediction model for the energy efficiency of air conditioner equipment in residential buildings, this embodiment expanded the number of questionnaires distributed to 460, basically covering all households in a typical residential area, and numerically meeting the Cochran formula's 95% confidence level and 5% allowable error requirements.

[0077] Based on the survey results, the typical air conditioning cooling and heating usage time patterns of bedroom air conditioners and living room air conditioners can be obtained by using the random prediction model of residential air conditioning usage behavior described in this invention. Figure 2 Based on the optimal cluster number judgment method described in the present invention, 3 typical bedroom cooling and 3 living room cooling usage schedules, as well as 4 bedroom heating and 5 living room heating usage schedules are clustered, which are represented by A, B, C, D, and E respectively. Using the residential air conditioning set temperature random prediction model described in the present invention, the probability distribution of air conditioning cooling and heating set temperatures for bedroom air conditioning and living room air conditioning can be obtained as follows: Figure 3 The random prediction model of energy efficiency of residential air-conditioning equipment described in the present invention can be used to obtain the probability distribution of energy efficiency of bedroom air-conditioning and living room air-conditioning equipment as follows: Figure 4 Using the prediction model of flexible adjustment willingness of residents in residential areas described in the present invention, the probability distribution of residents' willingness to participate in flexible adjustment and the probability distribution of their willingness to choose the degree of flexible adjustment can be obtained as follows: Figure 5 To set the peak and off-peak periods for flexible regulation, this embodiment makes assumptions about the peak and off-peak distribution of electricity prices based on the dynamic electricity prices obtained from literature research. The peak periods for residential electricity prices are set from 11:00 to 12:00 and from 18:00 to 20:00, and the off-peak periods are set from 23:00 to 5:00 and from 13:00 to 16:00.

[0078] Furthermore, the residential area air conditioning, cooling, and heating benchmark energy consumption prediction model of the present invention is used to predict the total air conditioning, cooling, and heating energy consumption during the benchmark daily peak period, as shown in Figures 6 and 7. Figure 7 This example also uses the root mean square error (RMSE) metric for the residential building complex air conditioning, cooling, and heating benchmark energy consumption prediction model. The unit area energy consumption obtained from the simulated air conditioning and cooling of bedrooms and living rooms in typical buildings was compared with the measured unit area energy consumption of residential air conditioning units to verify its accuracy. The results show that the CV-RMSE (Correction-Value-Residual-Separation) between the measured and simulated values ​​for bedrooms, living rooms, and the entire building combined is 6%, meeting the ASHRAE simulation accuracy requirement of a CV-RMSE of less than 15%.

[0079] By using the flexible adjustment strategy decision method for residential air conditioning of the present invention, the air conditioning set temperature after flexible adjustment is input into the energy consumption prediction model after flexible adjustment of the residential air conditioning equipment. The daily peak period total air conditioning cooling and heating energy consumption prediction result after flexible adjustment is obtained by Figure 6 and Figure 7 The gray dotted line in the figure shows the flexible regulation potential of air-conditioning equipment in a residential complex. The flexible regulation potential of air-conditioning equipment in the residential complex is predicted by combining the flexible regulation potential evaluation index. The hourly cooling and heating loads and load flexibility results of the residential complex air-conditioning on a typical day are shown in the figure below. Figure 8 As shown. In the figure, the base load of air conditioning in residential buildings is represented by a black solid line, the load after flexible adjustment of air conditioning in residential buildings is represented by a dotted line, and the flexibility of cooling and heating loads of air conditioning in residential buildings is represented by the gray area between the two lines. The flexibility of cooling and heating loads is approximately 10% and 28% of the base cooling and heating loads respectively. The total cooling and heating energy consumption and energy flexibility results of air conditioning in residential buildings during the cooling or heating season are shown in the figure below. Figure 9 The cooling and heating energy flexibility are approximately 20% and 29% of the baseline cooling and heating energy consumption, respectively. The total cooling and heating operating costs and flexibility benefits of the residential building complex during the cooling or heating season are shown in the figure below. Figure 10 The flexible benefits for cooling and heating are approximately 16% and 22% of the baseline cooling and heating operating costs, respectively.

[0080] Corresponding to the aforementioned embodiment of a method for predicting the potential for flexible regulation of air conditioning in a residential complex based on uncertainty, the present invention also provides an embodiment of a device for predicting the potential for flexible regulation of air conditioning in a residential complex based on uncertainty.

[0081] See also Figure 11An embodiment of the present invention provides a device for predicting the potential for flexible adjustment of air conditioning in a residential building complex based on uncertainty, comprising a memory and one or more processors, wherein the memory stores executable code, and when the processor executes the executable code, it is used to implement a method for predicting the potential for flexible adjustment of air conditioning in a residential building complex based on uncertainty in the above embodiment.

[0082] The embodiment of the uncertainty-based prediction device for the flexible adjustment potential of air conditioning in residential buildings provided by the present invention can be applied to any device with data processing capabilities, and the device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented through software, or through hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of any device with data processing capabilities in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory for execution. From the hardware level, if Figure 11 As shown, it is a hardware structure diagram of any device with data processing capability where the uncertainty-based residential building complex air conditioning flexible adjustment potential prediction device is located, except Figure 11 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0083] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0084] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present invention. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0085] An embodiment of the present invention also provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method for predicting the flexible adjustment potential of air conditioning in a residential building complex based on uncertainty in the above embodiment is implemented.

[0086] The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0087] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the uncertainty-based method for predicting the flexible adjustment potential of air conditioning in residential building complexes.

[0088] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any modifications and changes made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A method for predicting the flexible regulation potential of air conditioning in residential buildings based on uncertainty, characterized by: The specific steps of the prediction method are as follows: Step S1: Acquire flexible energy consumption information of air conditioners in residential buildings; obtain the air conditioner usage time and set temperature, air conditioner equipment performance, indoor and outdoor temperatures, and residents' flexible adjustment willingness to increase the cooling set temperature or decrease the heating set temperature for different residents in the residential area; extract typical characteristics of air conditioner usage time, calculate the distribution probability of typical air conditioner usage time patterns, set temperature, equipment energy efficiency, and residents' flexible adjustment willingness; and randomly predict the air conditioner usage time and set temperature, equipment energy efficiency, and flexible adjustment willingness of all residents in the residential area using the Monte Carlo method. Step S2: Construct a prediction model for the baseline energy consumption of air conditioning, cooling, and heating for the residential complex; based on the randomly predicted air conditioning usage time, set temperature, and equipment energy efficiency, complete the prediction of the baseline energy consumption of air conditioning, cooling, and heating for the residential complex through energy consumption simulation in the cooling and heating seasons; Step S3: Based on the randomly predicted flexible adjustment intentions of all residents in the residential area, a flexible adjustment strategy is determined for each air conditioner in each household during the cooling and heating seasons, and the set temperature of the air conditioner after flexible adjustment is obtained; Step S4: Constructing an energy consumption prediction model for the air conditioners in the residential complex after flexible adjustment; based on the set temperature after flexible adjustment obtained in step S3, the cooling and heating energy consumption predictions for the air conditioners in the residential complex after flexible adjustment are completed by simulating the energy consumption in the corresponding cooling and heating seasons; Step S5: Determine the flexible regulation potential evaluation index; based on the energy consumption prediction results of steps S2 and S4, predict the flexible regulation potential of the air conditioning of the residential building complex.

2. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S1, a random prediction model for air-conditioning usage behavior in residential buildings is constructed; typical characteristics of air-conditioning usage time in residential buildings are extracted and surveyed to obtain typical air-conditioning usage time patterns in the buildings; based on the probabilities of various typical patterns, the random air-conditioning usage time of all residents in the residential area is predicted separately using the Monte Carlo method.

3. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S1, a random prediction model for the air-conditioning set temperature of the residential building complex is constructed; based on the air-conditioning set temperature data collected from the survey, the distribution probability of the air-conditioning set temperature of the residential building complex is calculated; and then the corresponding set temperature is randomly assigned to each air-conditioning of each household through the Monte Carlo method.

4. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S1, a random prediction model for the energy efficiency of air-conditioning equipment in residential buildings is constructed; based on the energy efficiency data of air-conditioning equipment collected from the survey, the distribution probability of various energy efficiency levels of air-conditioning in residential buildings is calculated; and then the Monte Carlo method is used to randomly assign the corresponding energy efficiency level to each air-conditioning in each household.

5. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S1, a random prediction model of the flexible adjustment willingness of residents in a residential building complex is constructed; the residents' flexible adjustment willingness includes whether the air conditioner participates in flexible adjustment on that day and the flexible adjustment range of the air conditioner set temperature; the flexible adjustment range of the air conditioner set temperature is to increase the cooling set temperature or decrease the heating set temperature by 1°C to 4°C on the basis of the baseline set temperature; based on the statistics of residents' flexible adjustment willingness collected from the survey, the probability distribution of various types of flexible adjustment willingness is calculated; then, the corresponding flexible adjustment willingness is randomly assigned to each resident and each air conditioner through the Monte Carlo method to quantitatively characterize the uncertainty of the flexible adjustment willingness of residents in the residential area.

6. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S2, the building layout and thermal performance of the building envelope of the selected residential area are collected, and an energy consumption simulation model of the residential building complex is constructed using the EnergyPlus software; the air conditioning usage time, air conditioning set temperature and equipment energy efficiency randomly predicted in step S1 are used as the input parameters of the model, and the predicted benchmark energy consumption of air conditioning in the residential building is used as the output parameter of the model.

7. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1 is characterized in that: In step S3, based on the flexible adjustment willingness of the residents randomly predicted in step S1, the flexible adjustment strategy of the air conditioner on that day is determined; specifically, whether the flexible adjustment strategy of the air conditioner on that day is executed is determined based on the willingness of the air conditioner to participate in flexible adjustment on that day; if the flexible adjustment strategy is executed, the set temperature of the air conditioner after flexible adjustment is determined based on the willingness to select the flexible adjustment range of the air conditioner set temperature.

8. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 7 is characterized in that: In step S4, the input parameter is the set temperature of each air conditioner of each household in the residential building after flexible adjustment obtained by prediction in step S3; the output parameter is the energy consumption of the air conditioner in the residential building after flexible adjustment.

9. The uncertainty-based prediction method for flexible air conditioning regulation potential of residential complexes according to claim 1, characterized in that: In step S5, the load flexibility, energy flexibility and flexible benefits of the air-conditioning in the residential building complex after the flexible adjustment strategy is implemented are calculated to predict the flexible adjustment potential of the air-conditioning in the residential building complex; load flexibility refers to the change in operating load during the peak electricity price period before and after flexible adjustment; energy flexibility refers to the change in operating energy consumption during the peak electricity price period before and after flexible adjustment; flexible benefit refers to the change in operating cost during the adjustment period before and after flexible adjustment.

10. A device for predicting the potential of flexible air conditioning regulation in residential buildings based on uncertainty, comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that: When the processor executes the executable code, it implements a method for predicting the flexible adjustment potential of air conditioning in a residential complex based on uncertainty as described in any one of claims 1 to 9.

Images