Demand response performance key design parameter screening method based on joint simulation

Abstract

The invention provides a demand response performance key design parameter screening method based on joint simulation. The objective of the invention is to comprehensively screen numerous potential parameters influencing the building response performance and accurately identify the source of the uncertainty of the building air conditioner load demand response. The method comprises the following steps: constructing a demand response performance evaluation index system, and formulating a control strategy of a baseline scene and a response scene; constructing a building fine design sample database with statistical representation significance through a Latin hypercube stratified sampling method; building information is written into an EngyPlus file through Python, EngyPlus is cyclically called to collect simulation results under the two scenes, and random building response performance data is generated in combination with evaluation indexes; a PRCC sensitivity analysis method is adopted to research the influence effect of design variables on various evaluation indexes, and key design parameters influencing the response performance are accurately and rapidly screened out.

Description

Technical Field

[0001] The present invention relates to the field of architecture and is mainly used for building optimization design, specifically a method for screening key design parameters of building air-conditioning load demand response performance based on joint simulation. Background Art

[0002] As awareness of climate and environmental issues continues to grow, a global consensus is emerging to increase renewable energy generation to promote energy conservation and emissions reduction. China, a major consumer of electricity and investor in renewable energy, has been actively promoting demand response (DR) strategies to enhance the energy flexibility of its power infrastructure and mitigate the uncertainties associated with renewable energy generation. Air conditioning systems account for a significant portion of seasonal energy consumption, making them a prime candidate for implementing DR strategies in the building sector.

[0003] Passive thermal storage (passive thermal mass of buildings, such as the building envelope and furniture) is the primary response resource for air conditioning systems. Its primary principle is to alter cooling demand, thereby reshaping the electricity demand curve and providing demand response services. A building's cooling load is a key factor influencing subsequent air conditioning electrical loads. This cooling load is influenced by factors such as outdoor meteorological conditions, indoor thermal environment parameters, and the building's passive thermal mass (such as the building envelope and furniture).

[0004] The methods for studying the responsiveness of buildings mainly include simulation and experiment. When some scholars studied the relationship between energy flexibility and building components and technologies through simulation, they pointed out that different boundary conditions, such as outdoor temperature, infiltration, envelope quality and user behavior, will affect the load of the building and also affect the potential for energy flexibility. Some scholars evaluated the flexibility potential of two residential buildings with different insulation and airtightness levels through experiments. The results showed that the regulatory potential of thermal mass depends on many factors (insulation level, type of radiator, etc.) and changes over time (cold season and transition season). The time constant of buildings with poor insulation performance is relatively short, while the time constant of passive houses is longer.

[0005] In general, the significance of building design parameters in influencing demand response has been demonstrated in numerous studies. However, existing research primarily focuses on specific parameters, with few comprehensive studies examining the coupled effects of various building design parameters on air conditioning system demand response performance, failing to identify key influencing parameters. Furthermore, testing and demonstration in real buildings can be time-consuming and costly, making them unsuitable for large-scale studies. Summary of the Invention

[0006] In response to the above-mentioned existing problems and shortcomings, the present invention proposes a method for screening key design parameters of demand response performance based on joint simulation. The Latin hypercube sampling method is used to generate the parameter sample space. Python-EnergyPlus joint simulation is used in combination with evaluation indicators to generate multiple sets of random building response performance data. PRCC sensitivity analysis is then used to screen out parameters that have a significant impact on the response performance.

[0007] A method for screening key design parameters of demand response performance based on co-simulation, characterized by comprising the following steps:

[0008] S1. Build a demand response performance evaluation index system and formulate a demand response control strategy.

[0009] S2. Construct a sample database of architectural refined design that is statistically representative.

[0010] S3. Based on the sample database established in step S2, use Python to write the building information into EnergyPlus, call EnergyPlus cyclically to collect simulation results, and generate random building response performance data based on evaluation indicators.

[0011] S4. Based on the random building response performance data generated in step S3, a sensitivity analysis method is used to study the effects of design variables on various evaluation indicators, and to screen out key design parameters that affect response performance.

[0012] Furthermore, the step S1 of constructing a demand response performance evaluation index system and formulating a demand response control strategy specifically includes the following steps:

[0013] S11. The building demand response performance evaluation index mainly considers the maximum load reduction in the response phase and the cumulative load reduction in the response phase, and uses the maximum adjustable power ratio ( ) and the response capacity ratio (γ f ) is the indicator. and γ f The calculation formula is as follows:

[0014]

[0015]

[0016] in, is the maximum load reduction in the response phase, W; P b (t) is the baseline power at time t in baseline operation mode, W; E f is the cumulative load reduction during the response phase, kJ. 、P b (t) and Ef Obtained from EnergyPlus simulation.

[0017] S12. Develop two control strategies using the EnergyPlus thermostat. One is the baseline scenario, where the system operates normally; the other is the response scenario, which is achieved by increasing the indoor temperature set point.

[0018] S13. Definition of building physical model: Set boundary conditions in EnergyPlus software, including at least building size, layout and other control variable information.

[0019] Furthermore, the step S2 of constructing a statistically representative architectural refined design sample database specifically includes the following steps:

[0020] S21. Through theoretical analysis, select design variables that potentially affect the building's air conditioning load demand response. These variables include exterior wall U-value, exterior wall density, exterior wall specific heat, interior wall thermal conductivity, interior wall density, interior wall specific heat, carpeting presence, floor density, floor specific heat, roof U-value, roof density, roof specific heat, exterior window U-value, exterior window solar heat gain, exterior wall solar absorptivity, roof solar absorptivity, airtightness, interior furniture surface area, interior furniture density, interior furniture thermal conductivity, interior equipment emissivity, and building orientation. The limits of these design variables are subject to technological development and design standards.

[0021] S22. To cover as much potential design space as possible, define the variables as continuous uniform distribution. Use the Latin hypercube stratified sampling method to sample each group of variables n times and obtain the joint distribution matrix Input of the design variables, as shown in the following formula:

[0022]

[0023] The matrix has m rows and n columns, where m represents the number of samples, m = 10n. n represents the number of design variables. mn Indicates the value of the mth sampling of the nth design variable.

[0024] Furthermore, the generation of random building response performance data in step S3 specifically includes the following steps:

[0025] S31. The process of constructing the building performance output matrix is ​​as follows: Use Python to read each set of optimization variables in Input in turn and write them into the IDF file of EnergyPlus software. Then, perform the entire cooling period simulation of building performance in sequence according to the time step, and obtain the building performance output under two scenarios. The building performance output matrix is ​​shown as follows:

[0026]

[0027]

[0028] Output 1 represents the simulation results for the baseline scenario, and Output 2 represents the simulation results for the response scenario. y and z represent the building's power demand under different scenarios, including power demand and power consumption, respectively. p represents the type of building power demand and is 2 in both scenarios.

[0029] S32. Subtract the power demand in the baseline scenario from the power demand in the response scenario, and quantify the result according to the selected evaluation index to obtain the building's response performance RP.

[0030]

[0031] Furthermore, identifying key parameters using the sensitivity analysis method in step S4 specifically includes the following steps:

[0032] S41. The partial rank correlation coefficient (PRCC) is used as a sensitivity indicator to quantify the impact of building design variables on building response performance. The calculation formula is as follows:

[0033]

[0034]

[0035]

[0036] Here, x and y represent input and output respectively.

[0037] S42. Parameters with significant impact on response performance are screened through comparative analysis. PRCC values ​​range from -1 to 1, with the absolute value representing the weight of the impact. A larger absolute value indicates greater parameter importance. The sign indicates a positive or negative correlation between the two.

[0038] Compared with existing technologies, this invention offers the following advantages: it enables comprehensive screening of numerous potential parameters that influence building response performance, and utilizes Python-EnergyPlus co-simulation technology to effectively alleviate the time pressure associated with conducting simulation studies targeting a large number of parameters. Furthermore, this invention establishes a response performance rating index that fully accounts for the potential coupling effects between different design parameters. Through PRCC sensitivity analysis, this invention can accurately identify sources of demand response uncertainty, thus avoiding the tedious and time-consuming process of interpreting different parameters one by one. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 , a schematic diagram of a method for screening key design parameters of demand response performance based on example-based joint simulation of the present invention;

[0040] Figure 2 , the Python-EnergyPlus joint simulation flow chart of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the present invention provides a technical solution: a method for screening key design parameters of demand response performance based on joint simulation, comprising the following steps:

[0043] S1. Build a demand response performance evaluation index system and formulate a demand response control strategy.

[0044] S2. Construct a sample database of architectural refined design that is statistically representative.

[0045] S3. Based on the sample database established in step S2, use Python to write the building information into EnergyPlus, call EnergyPlus cyclically to collect simulation results, and generate random building response performance data based on evaluation indicators.

[0046] S4. Based on the random building response performance data generated in step S3, a sensitivity analysis method is used to study the effects of design variables on various evaluation indicators, and to screen out key design parameters that affect response performance.

[0047] Among them, the step S1 of constructing a scientific evaluation index system is the key to a reasonable analysis of the response. The formulation of response performance indicators includes the following steps:

[0048] S11. According to the response mechanism of air conditioning load, the building demand response performance evaluation index mainly considers the maximum load reduction amount and the cumulative load reduction amount in the response phase, respectively, with the maximum adjustable power ratio ( ) and the response capacity ratio (γ f ) is the indicator. and γ f The calculation formula is as follows:

[0049]

[0050]

[0051] in, is the maximum load reduction in the response phase, W; P b (t) is the baseline power at time t in baseline operation mode, W; E f is the cumulative load reduction during the response phase, kJ. 、P b (t) and E f Obtained from EnergyPlus simulation.

[0052] S12. Evaluation indicators are obtained based on a power demand baseline. Two control strategies are developed using the EnergyPlus thermostat. One is the baseline scenario, where the system operates normally; the other is the response scenario, which is achieved by increasing the indoor temperature set point.

[0053] S13. Setting of building physical model: Setting boundary conditions in EnergyPlus software, including at least control variable information such as building size and layout.

[0054] The step S2 constructs a statistically representative architectural refined design sample database, including the following steps:

[0055] S21. Through theoretical analysis, select design variables that have potential impacts on the building's air conditioning load demand response, including exterior wall U-value, exterior wall density, exterior wall specific heat, interior wall thermal conductivity, interior wall density, interior wall specific heat, presence or absence of carpet, floor density, floor specific heat, roof U-value, roof density, roof specific heat, exterior window U-value, exterior window solar heat gain, exterior wall solar absorptivity, roof solar absorptivity, airtightness, indoor furniture surface area, indoor furniture density, indoor furniture thermal conductivity, indoor equipment emissivity, and building orientation. The value boundaries of design variables are limited by technological development and design standards, and the specific range of design variables is set in accordance with GB55015-2021 (General Specification for Energy Conservation and Renewable Energy Utilization in Buildings).

[0056] S22. To cover as much potential design space as possible, define the variables as continuous uniform distribution. Use the Latin hypercube stratified sampling method to sample each group of variables n times and obtain the joint distribution matrix Input of the design variables, as shown in the following formula:

[0057]

[0058] The matrix has m rows and n columns, where m represents the number of samples, m = 10n. n represents the number of design variables. mn Indicates the value of the mth sampling of the nth design variable.

[0059] The step S3 of generating random building response performance data comprises the following steps:

[0060] S31. The process of constructing the building performance output matrix is ​​as follows: Use Python to read each set of optimization variables in Input in turn and write them into the IDF file of EnergyPlus software. Then, perform the entire cooling period simulation of building performance in sequence according to the time step, and obtain the building performance output under two scenarios. The building performance output matrix is ​​shown as follows:

[0061]

[0062]

[0063] Output 1 represents the simulation results for the baseline scenario, and Output 2 represents the simulation results for the response scenario. y and z represent the building's power demand under different scenarios, including power demand and power consumption, respectively. p represents the type of building power demand and is 2 in both scenarios.

[0064] S32. Subtract the power demand in the baseline scenario from the power demand in the response scenario, and quantify the result according to the selected evaluation index to obtain the building's response performance RP.

[0065]

[0066] The step S4 uses a sensitivity analysis method to identify key parameters, including the following steps:

[0067] S41. The partial rank correlation coefficient (PRCC) is used as a sensitivity indicator to quantify the impact of building design variables on building response performance. The calculation formula is as follows:

[0068]

[0069]

[0070]

[0071] Here, x and y represent input and output respectively.

[0072] S42. Parameters with significant impact on response performance are screened through comparative analysis. PRCC values ​​range from -1 to 1, with the absolute value representing the weight of the impact. A larger absolute value indicates greater parameter importance. The sign indicates a positive or negative correlation between the two.

[0073] The key design parameters that affect the building response performance are obtained through the above steps. When designing a building, according to the application scenario, the values ​​of the corresponding parameters can be appropriately increased or decreased to improve the building response performance.

Claims

1. A method for screening key design parameters of demand response performance based on joint simulation, characterized by: The following steps are involved: S1. Build a demand response performance evaluation index system and formulate a demand response control strategy; S2. Build a statistically representative sample database of architectural refinement designs; S3. Based on the sample database established in step S2, write the building information into EnergyPlus using Python, call EnergyPlus cyclically to collect simulation results, and generate random building response performance data in combination with evaluation indicators; S4. Based on the random building response performance data generated in step S3, a sensitivity analysis method is used to study the effects of design variables on various evaluation indicators, and to screen out key design parameters that affect response performance.

2. The method for screening key design parameters of demand response performance based on joint simulation according to claim 1, characterized in that: The first step is to build a demand response performance evaluation index system and specify the demand response control strategy. Specifically, the evaluation index is obtained based on the power demand baseline, and two different control strategies are formulated through the EnergyPlus thermostat: the baseline scenario and the response scenario.

3. The method for screening key design parameters of demand response performance based on joint simulation according to claim 1, characterized in that: The second step is to construct a statistically representative database of building refined design samples. Specifically, the database fully considers 22 potential design variables that affect the response to air-conditioning load demand, including exterior wall U-value, exterior wall density, exterior wall specific heat, interior wall thermal conductivity, interior wall density, interior wall specific heat, presence or absence of carpet, floor density, floor specific heat, roof U-value, roof density, roof specific heat, exterior window U-value, exterior window solar heat gain coefficient, exterior wall solar absorptivity, roof solar absorptivity, air tightness, indoor furniture surface area, indoor furniture density, indoor furniture thermal conductivity, indoor equipment emissivity, and building orientation.

4. The method for constructing a statistically representative architectural design sample database according to claim 3, wherein: The design variables are defined as continuous uniform distribution. Specifically, in order to cover as much potential design space as possible, the Latin hypercube stratified sampling method is used to extract multiple variables from each group, and the joint distribution matrix of the design variables is obtained as the input of the subsequent simulation.

5. The method for screening key design parameters of demand response performance based on joint simulation according to claim 1, characterized in that: The third step is to generate random building response performance data. Specifically, Python is used to read each set of optimization variables in the Input and write them to EnergyPlus. The entire cooling period simulation of the building performance is performed in sequence according to the time step to obtain the building performance output under the baseline scenario and the response scenario. EnergyPlus is called in a loop to collect the simulation results and generate random building response performance data, thus achieving a comprehensive screening of the many potential parameters that affect the building response performance.

6. The method for screening key design parameters of demand response performance based on joint simulation according to claim 1, characterized in that: The fourth step uses sensitivity analysis to study the impact of design variables on various evaluation indicators and identify key design parameters that influence response performance. Specifically, the partial rank correlation coefficient (PRCC) is used as a response performance rating indicator. The PRCC value ranges from -1 to 1, with its absolute value representing the magnitude of the impact and its sign indicating a positive or negative correlation. The results of the sensitivity analysis accurately identify the sources of demand response uncertainty and fully account for the potential coupling effects between different design parameters.

Images