Air conditioner load prediction model establishment method and system, and air conditioner load prediction method

Abstract

The application discloses an air conditioner load prediction model establishing method, system and air conditioner load prediction method, and the air conditioner load prediction model establishing method comprises the following steps: collecting air conditioner coefficients and external environment conditions of each air conditioner; establishing an aggregated air conditioner load prediction model according to the air conditioner coefficients and the external environment conditions of all single air conditioners; constructing a general battery model of an air conditioner aggregator according to air conditioner loads of an air conditioner aggregator of the aggregated air conditioner load prediction model; adjusting model constraint conditions of the general battery model of the air conditioner aggregator into uncertainty equations through air conditioner power utilization habits of users; and obtaining an air conditioner load prediction model considering uncertainty according to the uncertainty equations and the aggregated air conditioner load prediction model. The air conditioner power utilization habits of the users and the external environment conditions are considered in the air conditioner load prediction model, the uncertainty of the air conditioner load is reduced, and precise management and control of the air conditioner load are facilitated.

Description

Technical Field

[0001] This invention belongs to the field of flexible load modeling, and particularly relates to the method, system and method for establishing air conditioning load prediction model. Background Technology

[0002] With continuous economic development and rising living standards, the number of household appliances used is increasing, leading to record-breaking electricity loads and a persistently tense power supply and demand situation.

[0003] In the current situation, demand response for residential electricity consumption is of great significance. To fully explore the potential of air conditioning load demand response and better flexibly control it, a reasonable model needs to be established. Currently, the three main modeling methods for air conditioning load are simplified mathematical models, detailed physical models, and regression models based on historical regression data.

[0004] However, the simplified mathematical models, detailed physical models, and regression models based on historical regression data mentioned above do not take into account external interference factors such as weather and user behavior, which increases the uncertainty of air conditioning load and cannot intuitively depict the adjustability of air conditioning load, thus making it impossible to achieve precise control of air conditioning load. Summary of the Invention

[0005] This invention provides a method and system for establishing an air conditioning load forecasting model that takes into account uncertainty, in order to solve the problem that existing models do not consider external interference factors such as weather and user behavior, which leads to uncertainty in air conditioning load and makes it impossible to achieve accurate control of air conditioning load.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] Firstly, a method for establishing an air conditioning load forecasting model is provided, including:

[0008] Collect the air conditioning coefficient and external environmental conditions for each air conditioner;

[0009] Based on the air conditioning coefficients of all individual air conditioners and external environmental conditions, an aggregated air conditioning load prediction model is established.

[0010] Based on the air conditioning load of the air conditioning aggregator in the aggregated air conditioning load prediction model, a general battery model for the air conditioning aggregator is constructed.

[0011] The model constraints of the general battery model of air conditioner aggregators are adjusted into an uncertainty equation based on users' air conditioner power consumption habits.

[0012] Based on the uncertainty equation and the aggregated air conditioning load forecasting model, an air conditioning load forecasting model considering uncertainty is obtained.

[0013] Preferably, based on the air conditioning coefficients of all individual air conditioners and external environmental conditions, an aggregated air conditioning load prediction model is established, including:

[0014] Based on the air conditioning coefficient, the thermal resistance R, heat capacity C, energy efficiency coefficient COP, and rated power P of a single air conditioner are obtained. r,i and the on / off status of the HVAC system. i (t);

[0015] Indoor temperature θ is obtained based on external environmental conditions. i (t), outdoor temperature θ o and external interference factors ω i (t);

[0016] Establish the thermal dynamic balance equation for the i-th air conditioner , where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of thermal resistance R and heat capacity C, and b is the ratio of energy efficiency coefficient COP to heat capacity C;

[0017] The power variable s of a single air conditioner i (t)P i As dynamic power Dynamic power For continuous variable N i ∈[0,P r,i ], dynamic power Substituting into the thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation. ;

[0018] The expression for establishing the aggregated air conditioning load forecasting model is as follows:

[0019] ;

[0020] Among them, P tot (t) represents the air conditioning load of the air conditioning aggregator.

[0021] Preferably, based on the air conditioning load of the air conditioning aggregator in the aggregated air conditioning load prediction model, a general battery model for the air conditioning aggregator is constructed, including:

[0022] Obtain the HVAC set temperature for each air conditioner. Indoor temperature The range is ,in Adjust according to user comfort requirements;

[0023] The indoor temperature is controlled by air conditioning. Maintain at the set temperature At that time, Substituting into the dynamic equilibrium equation of deformation heat, we obtain the basic power of a single air conditioner. The expression is:

[0024] ;

[0025] Wherein, the R i The thermal resistance of the i-th air conditioner, the COP i Let be the energy efficiency coefficient of the i-th air conditioner;

[0026] Based on basic power and dynamic power To obtain the charging and discharging power The expression is: ;

[0027] Obtain the power status of a single air conditioner. State of electrical energy The expression is:

[0028] ;

[0029] Wherein, the C i Let i be the heat capacity of the i-th air conditioner;

[0030] The dynamic equilibrium equation of deformation heat and basic power The expression and state of electrical energy By combining the expressions, we get , where α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner;

[0031] After discretization, the expression for the preliminary battery model is obtained. , where α i =1-Δt / C i R i The Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let be the electrical energy state of the i-th air conditioner at time k+1, and Δt be the time change between time k and time k+1;

[0032] The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k) yields the expression for the charging and discharging power of the air conditioning aggregator. ;

[0033] The formula for calculating the total power of an air conditioning aggregator is as follows:

[0034] ;

[0035] The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioning aggregator. b (k) yields the upper and lower boundary values ​​of the charging and discharging power of the air conditioner aggregator:

[0036] ;

[0037] The formulas for obtaining the conditions for minimizing and maximizing the expected temperature deviation of the i-th air conditioner are as follows:

[0038] ;

[0039] The upper and lower boundary values ​​of the aggregation power of the air conditioning aggregator are calculated based on the aforementioned conditional formula. and Combined with the expression of the preliminary battery model The upper and lower boundary values ​​of the electrical energy state of the air conditioning aggregator are obtained:

[0040] ;

[0041] Among them, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k;

[0042] Based on the preliminary battery model expression, the upper and lower boundary values ​​of charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression for the general battery model of the air conditioner aggregator is obtained:

[0043] .

[0044] Preferably, the model constraints of the general battery model of the air conditioner aggregator are adjusted into uncertainty equations based on users' air conditioner electricity consumption habits, including:

[0045] By analyzing users' air conditioning usage habits, we can obtain the user's response intention for the i-th air conditioner. ;

[0046] Based on user response intentions The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge and discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model for air conditioner aggregators:

[0047] ;

[0048] in, Let k be the value of the user's desired response at time k.

[0049] Preferably, the user's willingness to respond to the i-th air conditioner is obtained by analyzing the user's air conditioning electricity consumption habits. ,include:

[0050] Define the user's air conditioning electricity consumption habits and user response willingness for the i-th air conditioner as follows: User response willingness The expression is:

[0051] ;

[0052] Where, ρ i It's user sensitivity. For the incentive level, λ i The impact of weather factors on users, k is a positive real number. λ represents the difference between indoor temperature and standard comfort temperature, and λ0 represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum.

[0053] Obtaining user response willingness The membership function expression is:

[0054] ;

[0055] in, This represents the upper limit of a user's willingness to respond. This represents the lower limit of the user's willingness to respond. This represents the average level of user willingness to respond. , and The expression is:

[0056] ;

[0057] User response intentions Represented as a fuzzy random function:

[0058] ;

[0059] ρ i It follows a normal distribution. and ρ i The mean and standard deviation, For fuzzy variables, It is a random variable.

[0060] Secondly, a system for establishing an air conditioning load forecasting model is provided, including:

[0061] The data acquisition module is used to collect the air conditioning coefficient and external environmental conditions of each air conditioner.

[0062] The first modeling module is used to establish an aggregated air conditioning load prediction model based on the air conditioning coefficients of all individual air conditioners and external environmental conditions.

[0063] The second modeling module is used to construct a general battery model for air conditioning aggregators based on the air conditioning load of the aggregator in the aggregated air conditioning load prediction model.

[0064] The uncertainty calculation module is used to adjust the model constraints of the general battery model of air conditioner aggregators into uncertainty equations based on users' air conditioner power consumption habits.

[0065] The uncertainty modeling module is used to obtain an air conditioning load forecasting model that takes uncertainty into account, based on the uncertainty equation and the aggregated air conditioning load forecasting model.

[0066] Preferably, the first modeling module is specifically used to execute:

[0067] Based on the aforementioned air conditioning coefficient, the thermal resistance R, heat capacity C, energy efficiency coefficient COP, and rated power P of a single air conditioner are obtained. r,i and the on / off status of the HVAC system. i (t);

[0068] The indoor temperature θ is obtained based on the external environmental conditions. i (t), outdoor temperature θ o and external interference factors ω i (t);

[0069] Establish the thermal dynamic balance equation for the i-th air conditioner Where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of the thermal resistance R and the heat capacity C, and b is the ratio of the coefficient of performance (COP) to the heat capacity C;

[0070] The power variable s of a single air conditioner i (t)P i As dynamic power The dynamic power For continuous variable N i ∈[0,P r,i ], to the dynamic power Substituting into the aforementioned thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation. ;

[0071] The expression for establishing the aggregated air conditioning load forecasting model is as follows:

[0072] ;

[0073] Wherein, P tot (t) represents the air conditioning load of the air conditioning aggregator.

[0074] Preferably, the second modeling module is specifically used for execution:

[0075] Obtain the HVAC set temperature for each air conditioner. Indoor temperature The range is Δθ is adjusted according to user comfort requirements;

[0076] The indoor temperature is controlled by air conditioning. Maintain at the set temperature At that time, Substituting into the dynamic equilibrium equation of deformation heat, we obtain the basic power of a single air conditioner. The expression is:

[0077] ;

[0078] Among them, R i Let COP be the thermal resistance of the i-th air conditioner. i Let be the energy efficiency coefficient of the i-th air conditioner;

[0079] Based on basic power and dynamic power To obtain the charging and discharging power The expression is: ;

[0080] Obtain the power status of a single air conditioner. State of electrical energy The expression is:

[0081] ;

[0082] Among them, C i Let i be the heat capacity of the i-th air conditioner;

[0083] The dynamic equilibrium equation of deformation heat and basic power The expression and state of electrical energy By combining the expressions, we get , where α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner;

[0084] After discretization, the expression for the preliminary battery model is obtained. , where α i =1-Δt / C i R i , Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let be the electrical energy state of the i-th air conditioner at time k+1, and Δt be the time change between time k and time k+1;

[0085] The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k) yields the expression for the charging and discharging power of the air conditioning aggregator. ;

[0086] The formula for calculating the total power of an air conditioning aggregator is as follows:

[0087] ;

[0088] The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioning aggregator. b (k) yields the upper and lower boundary values ​​of the charging and discharging power of the air conditioner aggregator:

[0089] ;

[0090] The formulas for obtaining the conditions for minimizing and maximizing the expected temperature deviation of the i-th air conditioner are as follows:

[0091] ;

[0092] The upper and lower boundary values ​​of the aggregation power of the air conditioning aggregator are calculated based on the conditional formula. and Combined with the expression of the preliminary battery model The upper and lower boundary values ​​of the electrical energy state of the air conditioning aggregator are obtained:

[0093] ;

[0094] Among them, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k.

[0095] Based on the preliminary battery model expression, the upper and lower boundary values ​​of charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression for the general battery model of the air conditioner aggregator is obtained:

[0096] .

[0097] Optional, an uncertainty calculation module, specifically used for execution:

[0098] Define the user's air conditioning electricity consumption habits and user response willingness for the i-th air conditioner as follows: User response willingness The expression is:

[0099] ;

[0100] Where, ρ i It's user sensitivity. For the incentive level, λ i The impact of weather factors on users, k is a positive real number. λ represents the difference between indoor temperature and standard comfort temperature, and λ0 represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum.

[0101] Obtaining user response willingness The membership function expression is:

[0102] ;

[0103] in, This represents the upper limit of a user's willingness to respond. This represents the lower limit of the user's willingness to respond. This represents the average level of user willingness to respond. , and The expression is:

[0104] ;

[0105] User response intentions Represented as a fuzzy random function:

[0106] ;

[0107] ρ i It follows a normal distribution. and ρ i The mean and standard deviation, For fuzzy variables, It is a random variable;

[0108] Based on user response intentions The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge and discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model for air conditioner aggregators:

[0109] ;

[0110] Let be the value of the user's willingness to respond at time k.

[0111] Thirdly, a method for predicting air conditioning load is provided, including:

[0112] Collect the daytime temperature data, predict the daytime temperature data based on the daytime temperature data, and predict the predicted temperature value for the next time based on the daytime temperature data.

[0113] Based on the previous day's temperature data, the intraday temperature data, the predicted temperature value, and the air conditioning load prediction model considering uncertainties generated in the first aspect's air conditioning load prediction model establishment method, the air conditioning load at the next moment is predicted.

[0114] The beneficial effects achieved by this invention are as follows:

[0115] The system collects the air conditioning coefficient (ACF) of each air conditioner and external environmental conditions. Based on the ACF of all individual air conditioners and external environmental conditions, an aggregated air conditioning load prediction model is established. Based on the air conditioning load of the aggregated air conditioning load prediction model, a general battery model for the aggregated air conditioning load is constructed. The constraints of the general battery model are adjusted to an uncertainty equation based on user air conditioning usage habits. Based on the uncertainty equation and the aggregated air conditioning load prediction model, an air conditioning load prediction model considering uncertainty is obtained. By incorporating user air conditioning usage habits and external environmental conditions into the air conditioning load prediction model, the uncertainty of the air conditioning load is reduced, which helps to achieve precise control of the air conditioning load. Attached Figure Description

[0116] Figure 1 This is a flowchart of the method for establishing the air conditioning load prediction model of the present invention;

[0117] Figure 2 The structural diagram of the system for establishing the air conditioning load prediction model of this invention is shown. Detailed Implementation

[0118] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0119] like Figure 1 As shown, this embodiment of the invention provides a method for establishing an air conditioning load prediction model, including the following steps:

[0120] 101. Collect the air conditioning coefficient and external environmental conditions for each air conditioner;

[0121] The air conditioning coefficient is determined by factors such as the performance of the air conditioning product, the manufacturing process, and the parameters of its internal components. External environmental conditions can include factors such as weather and temperature.

[0122] 102. Based on the air conditioning coefficient of all individual air conditioners and the external environmental conditions, establish an aggregated air conditioning load prediction model.

[0123] The specific process of establishing the aggregated air conditioning load prediction model is as follows:

[0124] Based on the air conditioning coefficient, the thermal resistance R, heat capacity C, coefficient of performance (COP), and rated power P of a single air conditioner are obtained. r,i and the on / off status of the air conditioning system (Heating, Ventilation and Air Conditioning, HVAC). i (t), switch state s i The value of (t) is 0 or 1, where 1 is on and 0 is off;

[0125] Indoor temperature θ is obtained based on external environmental conditions. i (t), outdoor temperature θ o and external interference factors ω i (t);

[0126] Assuming there are a total of n air conditioners, the thermal dynamic balance equation for the i-th air conditioner is established as follows:

[0127] ;

[0128] Where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of thermal resistance R and heat capacity C, and b is the ratio of energy efficiency coefficient COP to heat capacity C;

[0129] The power variable s of a single air conditioner i (t)P i As dynamic power Dynamic power For continuous variable N i ∈[0,P r,i ], dynamic power Substituting into the thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation:

[0130] ;

[0131] The expression for establishing the aggregated air conditioning load forecasting model is as follows:

[0132] ;

[0133] Among them, P tot (t) represents the air conditioning load of the air conditioning aggregator.

[0134] 103. Based on the air conditioning load of the air conditioning aggregator in the aggregated air conditioning load prediction model, construct a general battery model for the air conditioning aggregator.

[0135] The detailed construction process in this step is as follows:

[0136] Obtain the HVAC set temperature for each air conditioner. Indoor temperature θ i (t) range is Δθ is adjusted according to user comfort requirements;

[0137] The indoor temperature is controlled by air conditioning. Maintain at the set temperature At that time, Substituting into the dynamic equilibrium equation of deformation heat, we obtain the basic power of a single air conditioner. The expression is:

[0138] ;

[0139] Among them, R i Let COP be the thermal resistance of the i-th air conditioner. i Let be the energy efficiency coefficient of the i-th air conditioner;

[0140] From the perspective of the power grid, if the air conditioner's power output is higher than its base power, it can be considered as a battery "charging"; if the air conditioner's power output is lower than its base power, it can be considered as a battery "discharging". Air conditioners need to maintain indoor temperatures within a certain range. In cooling mode, when the indoor temperature of all rooms reaches the upper limit, the air conditioner aggregator will stop discharging, and the corresponding state of energy is "depleted". Similarly, when the indoor temperature of all rooms reaches the lower limit, the air conditioner aggregator will stop charging, and the corresponding state of energy is "fully charged". The charging and discharging power of the air conditioner aggregator is essentially the difference between its dynamic power and base power, and the state of energy essentially represents the degree to which the room temperature deviates from the set temperature. For a single air conditioner, based on the base power... and dynamic power To obtain the charging and discharging power The expression is: ;

[0141] Obtain the power status of a single air conditioner. State of electrical energy The expression is:

[0142] C i Let i be the heat capacity of the i-th air conditioner;

[0143] The dynamic equilibrium equation of deformation heat and basic power The expression and state of electrical energy By combining the expressions, we get , where α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner;

[0144] After discretization, the expression for the preliminary battery model is obtained. , where α i =1-Δt / C i R i ; Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let be the electrical energy state of the i-th air conditioner at time k+1, and Δt be the time change between time k and time k+1;

[0145] To determine the upper and lower boundaries of charging and discharging power and state of charge, the basic power of the air conditioner aggregator must first be determined. The basic power P of the air conditioner aggregator at time k is then determined by calculating the power required for the air conditioner to maintain the set temperature. b (k) uses the predicted outdoor temperature of the next day to simulate the basic power of the air conditioning aggregator;

[0146] The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k) yields the expression for the charging and discharging power of the air conditioning aggregator. ;

[0147] The formula for calculating the total power of an air conditioning aggregator is as follows:

[0148] ;

[0149] The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioning aggregator. b (k) yields the upper and lower boundary values ​​of the charging and discharging power of the air conditioner aggregator:

[0150] ;

[0151] It should be noted that the upper and lower boundary values ​​of the state of energy (SEO) for air conditioner aggregators are theoretically the SEO values ​​when all rooms are at the lower and upper limits of the temperature dead zone. However, in reality, these two scenarios are rarely encountered. Therefore, we choose to calculate the SEO values ​​when the expected temperature deviation of all rooms is minimized and maximized to obtain its upper and lower boundaries, and then obtain the conditional formulas for the expected minimum and maximum expected temperature deviation of the i-th air conditioner:

[0152] ;

[0153] The upper and lower boundary values ​​of the aggregation power of the air conditioning aggregator are calculated based on the conditional formula. and Combined with the expression of the preliminary battery model The upper and lower boundary values ​​of the electrical energy state of the air conditioning aggregator are obtained:

[0154] ;

[0155] Among them, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k.

[0156] Based on the preliminary battery model expression, the upper and lower boundary values ​​of charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression for the general battery model of the air conditioner aggregator is obtained:

[0157] .

[0158] Based on the above calculations, the parameters in the general battery model for air conditioner aggregators are shown in Table 1 below.

[0159] Table 1. Parameters of the general battery model for air conditioner aggregators

[0160]

[0161] 104. By adjusting the model constraints of the general battery model of air conditioner aggregators into an uncertainty equation based on users' air conditioner electricity consumption habits;

[0162] Users' air conditioning usage habits can be specifically reflected in their behavior. However, this behavior is influenced by various factors such as weather, making it difficult to predict accurately and increasing uncertainty. This embodiment primarily considers the susceptibility of the upper and lower boundary values ​​of charging and discharging power and the upper and lower boundary values ​​of the state of energy to uncertainty. Therefore, by analyzing users' air conditioning usage habits, the user's willingness to respond to the i-th air conditioner is obtained. ;

[0163] Based on user response intentions The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge and discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model for air conditioner aggregators:

[0164] ;

[0165] in, Let be the value of the user's willingness to respond at time k.

[0166] 105. Based on the uncertainty equation and the aggregated air conditioning load prediction model, an air conditioning load prediction model considering uncertainty is obtained.

[0167] The implementation principle of this invention is as follows:

[0168] The system collects the air conditioning coefficient (ACF) of each air conditioner and external environmental conditions. Based on the ACF of all individual air conditioners and external environmental conditions, an aggregated air conditioning load prediction model is established. Based on the air conditioning load of the aggregated air conditioning load prediction model, a general battery model for the aggregated air conditioning load is constructed. The constraints of the general battery model are adjusted to an uncertainty equation based on user air conditioning usage habits. Based on the uncertainty equation and the aggregated air conditioning load prediction model, an air conditioning load prediction model considering uncertainty is obtained. By incorporating user air conditioning usage habits and external environmental conditions into the air conditioning load prediction model, the uncertainty of the air conditioning load is reduced, which helps to achieve precise control of the air conditioning load.

[0169] Optional, in the above Figure 1 In step 104 of the embodiment shown, the user's willingness to respond to the i-th air conditioner is obtained by analyzing the user's air conditioning electricity consumption habits. ,include:

[0170] Define the user's air conditioning electricity consumption habits and user response willingness for the i-th air conditioner as follows: User response willingness The expression is:

[0171] ;

[0172] Where, ρ i It is user sensitivity, ∆m i For the incentive level, λ i The impact of weather factors on users, k is a positive real number, ∆T i λ represents the difference between indoor temperature and standard comfort temperature, and λ0 represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum.

[0173] Obtaining user response willingness The membership function expression is:

[0174] ;

[0175] in, This represents the upper limit of a user's willingness to respond. This represents the lower limit of the user's willingness to respond. This represents the average level of user willingness to respond. , and The expression is:

[0176] ;

[0177] User response intentions Represented as a fuzzy random function:

[0178] ;

[0179] ρ i It follows a normal distribution. and ρ i The mean and standard deviation, For fuzzy variables, It is a random variable;

[0180] User response intentions As an uncertainty.

[0181] The principle of this invention is as follows: the impact of weather factors on user behavior and the uncertainty brought about by user habits are quantified by the user's willingness to respond, and the application of fuzzy random functions makes the quantification of uncertainty more accurate, thereby further improving the accuracy of air conditioning load.

[0182] In the above embodiments, the method for establishing an air conditioning load forecasting model considering uncertainty provided by the present invention has been described. The following embodiments will specifically describe the system for establishing an air conditioning load forecasting model considering uncertainty, such as… Figure 2 As shown, it includes:

[0183] The data acquisition module 201 is used to collect the air conditioning coefficient and external environmental conditions of each air conditioner.

[0184] The first modeling module 202 is used to establish an aggregated air conditioning load prediction model based on the air conditioning coefficients of all individual air conditioners and external environmental conditions.

[0185] The second modeling module 203 is used to construct a general battery model for air conditioning aggregators based on the air conditioning load of the air conditioning aggregators in the aggregated air conditioning load prediction model.

[0186] Uncertainty calculation module 204 is used to adjust the model constraints of the general battery model of air conditioner aggregators into uncertainty equations based on users' air conditioner electricity consumption habits;

[0187] Uncertainty modeling module 205 is used to obtain an air conditioning load forecasting model that takes uncertainty into account, based on the uncertainty equation and the aggregated air conditioning load forecasting model.

[0188] The implementation principle of this invention is as follows:

[0189] The data acquisition module 201 collects the air conditioning coefficient and external environmental conditions for each air conditioner. The first modeling module 202 establishes an aggregated air conditioning load prediction model based on the air conditioning coefficients and external environmental conditions of all individual air conditioners. The second modeling module 203 constructs a general battery model for the air conditioning aggregator based on the air conditioning load of the aggregator in the aggregated air conditioning load prediction model. The uncertainty calculation module 204 adjusts the model constraints of the general battery model of the air conditioning aggregator into an uncertainty equation based on users' air conditioning electricity consumption habits. The uncertainty modeling module 205 obtains an air conditioning load prediction model that considers uncertainty based on the uncertainty equation and the aggregated air conditioning load prediction model. By incorporating users' air conditioning electricity consumption habits and external environmental conditions into the prediction model that takes into account the air conditioning load, the uncertainty of the air conditioning load is reduced, which helps to achieve precise control of the air conditioning load.

[0190] In summary Figure 2 In the embodiments shown, and in some preferred embodiments of the present invention, the first modeling module 202 is specifically used to perform the following processes:

[0191] Based on the aforementioned air conditioning coefficient, the thermal resistance R, heat capacity C, energy efficiency coefficient COP, and rated power P of a single air conditioner are obtained. r,i and the on / off status of the HVAC system. i (t);

[0192] The indoor temperature θ is obtained based on the external environmental conditions. i (t), outdoor temperature θ o and external interference factors ω i (t);

[0193] Establish the thermal dynamic balance equation for the i-th air conditioner Where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of the thermal resistance R and the heat capacity C, and b is the ratio of the coefficient of performance (COP) to the heat capacity C;

[0194] The power variable s of a single air conditioner i (t)P i As dynamic power The dynamic power For continuous variable N i ∈[0,P r,i ], to the dynamic power Substituting into the aforementioned thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation. ;

[0195] The expression for establishing the aggregated air conditioning load forecasting model is as follows:

[0196] ;

[0197] Wherein, P tot (t) represents the air conditioning load of the air conditioning aggregator.

[0198] In summary Figure 2 In the embodiments shown, and in some preferred embodiments of the present invention, the second modeling module 203 is specifically used to perform the following processes:

[0199] Obtain the HVAC set temperature for each air conditioner. Indoor temperature θ i (t) range is Δθ is adjusted according to user comfort requirements;

[0200] The indoor temperature is controlled by air conditioning. Maintain at the set temperature At that time, Substituting into the dynamic equilibrium equation of deformation heat, we obtain the basic power of a single air conditioner. The expression is:

[0201] ;

[0202] Among them, R i Let COP be the thermal resistance of the i-th air conditioner. i Let be the energy efficiency coefficient of the i-th air conditioner;

[0203] Based on basic power and dynamic power To obtain the charging and discharging power The expression is: ;

[0204] Obtain the power status of a single air conditioner. State of electrical energy The expression is:

[0205] ;

[0206] Among them, C i Let i be the heat capacity of the i-th air conditioner;

[0207] The dynamic equilibrium equation of deformation heat and basic power The expression and state of electrical energy By combining the expressions, we get , where α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner;

[0208] After discretization, the expression for the preliminary battery model is obtained. , where α i=1-Δt / C i R i , Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let be the electrical energy state of the i-th air conditioner at time k+1, and Δt be the time change between time k and time k+1;

[0209] The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k) yields the expression for the charging and discharging power of the air conditioning aggregator. ;

[0210] The formula for calculating the total power of an air conditioning aggregator is as follows:

[0211] ;

[0212] The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioning aggregator. b (k) yields the upper and lower boundary values ​​of the charging and discharging power of the air conditioner aggregator:

[0213] ;

[0214] The formulas for obtaining the conditions for minimizing and maximizing the expected temperature deviation of the i-th air conditioner are as follows:

[0215] ;

[0216] The upper and lower boundary values ​​of the aggregation power of the air conditioning aggregator are calculated based on the conditional formula. and Combined with the expression of the preliminary battery model The upper and lower boundary values ​​of the electrical energy state of the air conditioning aggregator are obtained:

[0217] ;

[0218] Among them, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k.

[0219] Based on the preliminary battery model expression, the upper and lower boundary values ​​of charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression for the general battery model of the air conditioner aggregator is obtained:

[0220] .

[0221] In summary Figure 2In the embodiments shown, and in some preferred embodiments of the present invention, the uncertainty calculation module 204 is specifically used to perform the following processes:

[0222] Define the user's air conditioning electricity consumption habits and user response willingness for the i-th air conditioner as follows: User response willingness The expression is:

[0223] ;

[0224] Where, ρ i It's user sensitivity. For the incentive level, λ i The impact of weather factors on users, k is a positive real number. λ represents the difference between indoor temperature and standard comfort temperature, and λ0 represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum.

[0225] Obtaining user response willingness The membership function expression is:

[0226] ;

[0227] in, This represents the upper limit of a user's willingness to respond. This represents the lower limit of the user's willingness to respond. This represents the average level of user willingness to respond. , and The expression is:

[0228] ;

[0229] User response intentions Represented as a fuzzy random function:

[0230] ;

[0231] ρ i It follows a normal distribution. and ρ i The mean and standard deviation, For fuzzy variables, It is a random variable;

[0232] Based on user response intentions The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge and discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model for air conditioner aggregators:

[0233] ;

[0234] Let be the value of the user's willingness to respond at time k.

[0235] In the above embodiments of the air conditioning load forecasting model establishment method and system, after the air conditioning load forecasting model considering uncertainties is generated, it is used for air conditioning load forecasting. The process of air conditioning load forecasting is described below through embodiments:

[0236] Collect the daytime temperature data, predict the daytime temperature data based on the daytime temperature data, and predict the predicted temperature value for the next time based on the daytime temperature data.

[0237] Based on the daytime temperature data, the intraday temperature data, the predicted temperature value, and the air conditioning load prediction model considering uncertainties, the model outputs the predicted air conditioning load for the next time moment.

[0238] Because the air conditioning load forecasting model takes into account users' air conditioning electricity consumption habits and external environmental conditions when predicting the air conditioning load at the next moment, the forecasting of air conditioning load is more accurate.

[0239] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0240] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0241] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0242] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0243] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A method for establishing an air conditioning load forecasting model, characterized in that, include: The air conditioning coefficient (ACC) and external environmental conditions for each air conditioner are collected. The ACC includes thermal resistance R, heat capacity C, coefficient of performance (COP), and rated power P. r,i and the on / off status of the HVAC system. i (t), the external environmental conditions include indoor temperature θ i (t), outdoor temperature θ o and external interference factors ω i (t); Establish the thermal dynamic balance equation for the i-th air conditioner Where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of the thermal resistance R and the heat capacity C, and b is the ratio of the energy efficiency coefficient COP to the heat capacity C; The power variable s of a single air conditioner i (t)P i As dynamic power The dynamic power For continuous variable N i ∈[0, P r,i ], to the dynamic power Substituting into the aforementioned thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation. ; The expression for establishing the aggregated air conditioning load forecasting model is as follows: ； Wherein, P tot (t) represents the air conditioning load of the air conditioning aggregator; Based on the air conditioning load of the air conditioning aggregator in the aggregated air conditioning load prediction model, a general battery model for the air conditioning aggregator is constructed. By analyzing users' air conditioning usage habits, we can obtain the user's response intention for the i-th air conditioner. ; Based on the user's response intention The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge / discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model of the air conditioner aggregator. ； Among them, the For the charging and discharging power at time k, the For the The lower boundary value, the For the The upper boundary value, the For the state of electrical energy at time k, the For the The lower boundary value, the For the The upper boundary value, the The user's response intention at time k The value of ; Based on the uncertainty equation and the aggregated air conditioning load prediction model, an air conditioning load prediction model considering uncertainty is obtained.

2. The method for establishing an air conditioning load forecasting model according to claim 1, characterized in that, The step of constructing a general battery model for air conditioning aggregators based on the air conditioning load of the aggregator in the aggregated air conditioning load prediction model includes: Obtain the set temperature of the HVAC for each air conditioner. Indoor temperature θ i (t) ranges Δθ is adjusted according to user comfort requirements; The indoor temperature θ is controlled by air conditioning. i (t) Maintain at the set temperature At that time, Substituting these values ​​into the aforementioned dynamic equilibrium equation for heat deformation, we obtain the basic power of a single air conditioner. The expression is: ； Wherein, the R i The thermal resistance of the i-th air conditioner, the COP i Let be the energy efficiency coefficient of the i-th air conditioner; According to the basic power and the dynamic power To obtain the charging and discharging power The expression is: ; Obtain the power status of a single air conditioner. The state of electrical energy The expression is: ； Wherein, the C i Let the heat capacity of the i-th air conditioner be denoted as . The deformation thermal dynamic equilibrium equation and the basic power The expression and the state of electrical energy By combining the expressions, we get Wherein α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner; After discretization, the expression for the preliminary battery model is obtained. Wherein α i =1-Δt / C i R i The Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let Δt be the electrical energy state of the i-th air conditioner at time k+1, and let Δt be the time change between time k and time k+1. The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k), to obtain the expression ; The formula for calculating the total power of an air conditioning aggregator is as follows: ； The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioner aggregator. b (k), to obtain the Upper and lower boundary values: ； The formulas for obtaining the conditions for minimizing and maximizing the expected temperature deviation of the i-th air conditioner are as follows: ； The upper and lower boundary values ​​of the aggregation power of the air conditioner aggregator are calculated according to the aforementioned conditional formula. and Combined with the expression of the preliminary battery model , to obtain the Upper and lower boundary values: ； Wherein, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k; Based on the expression of the preliminary battery model, the upper and lower boundary values ​​of the charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression of the general battery model for air conditioner aggregators is obtained: 。 3. The method for establishing an air conditioning load prediction model according to claim 1, characterized in that, The user's response intention for the i-th air conditioner is obtained by analyzing the user's air conditioning usage habits. ,include: Define the user's willingness to respond to the user's air conditioning electricity consumption habits for the i-th air conditioner as... The user's willingness to respond The expression is: ； Wherein, the ρ i It is user sensitivity, the ∆m i As the incentive level, the λ i Regarding the impact of weather factors on users, the Where k is a positive real number, and ∆T i The difference between indoor temperature and standard comfort temperature is λ0, which represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum. Obtain the user's response intention The membership function expression is: ； Among them, the The upper limit of the user's willingness to respond, the As the lower limit of the user's willingness to respond, the The average value of user response intention, the The above and the aforementioned The expression is: ； The user's response intention Represented as a fuzzy random function: ； The ρ i Satisfying a normal distribution, the and stated respectively ρ i The mean and standard deviation, the As a fuzzy variable, the It is a random variable.

4. A system for establishing an air conditioning load prediction model, characterized in that, include: The data acquisition module is used to collect the air conditioning coefficient and external environmental conditions for each air conditioner. The air conditioning coefficient includes thermal resistance R, heat capacity C, coefficient of performance (COP), and rated power P. r,i and the on / off status of the HVAC system. i (t), the external environmental conditions include indoor temperature θ i (t), outdoor temperature θ o and external interference factors ω i (t); The first modeling module is used to establish an aggregated air conditioning load prediction model based on the air conditioning coefficients of all individual air conditioners and the external environmental conditions. The second modeling module is used to construct a general battery model for air conditioning aggregators based on the air conditioning load of the air conditioning aggregators in the aggregated air conditioning load prediction model. The uncertainty calculation module is used to determine the user's willingness to respond to the i-th air conditioner by analyzing the user's air conditioning electricity consumption habits. Based on the user's response intention The uncertainty equation is derived by considering the impact of the upper and lower boundary values ​​of charge / discharge power and the upper and lower boundary values ​​of state of energy in the model constraints of the general battery model of the air conditioner aggregator. ； Among them, the For the charging and discharging power at time k, the For the The lower boundary value, the For the The upper boundary value, the For the state of electrical energy at time k, the For the The lower boundary value, the For the The upper boundary value, the The user's response intention at time k The value of ; An uncertainty modeling module is used to obtain an air conditioning load forecasting model that takes uncertainty into account, based on the uncertainty equation and the aggregated air conditioning load forecasting model. The first modeling module is specifically used to execute: Establish the thermal dynamic balance equation for the i-th air conditioner Where i is a positive integer greater than or equal to 1 and less than or equal to n, a is the product of the thermal resistance R and the heat capacity C, and b is the ratio of the energy efficiency coefficient COP to the heat capacity C; The power variable s of a single air conditioner i (t)P i As dynamic power The dynamic power For continuous variable N i ∈[0, P r,i ], to the dynamic power Substituting into the aforementioned thermal dynamic equilibrium equation, we obtain the deformation thermal dynamic equilibrium equation. ; The expression for establishing the aggregated air conditioning load forecasting model is as follows: ； Wherein, P tot (t) represents the air conditioning load of the air conditioning aggregator.

5. The air conditioning load prediction model establishment system according to claim 4, characterized in that, The second modeling module is specifically used to execute: Obtain the set temperature of the HVAC for each air conditioner. Indoor temperature The range is Δθ is adjusted according to user comfort requirements; The indoor temperature is controlled by air conditioning. Maintain at the set temperature At that time, Substituting these values ​​into the aforementioned dynamic equilibrium equation for heat deformation, we obtain the basic power of a single air conditioner. The expression is: ； Wherein, the R i The thermal resistance of the i-th air conditioner, the COP i Let be the energy efficiency coefficient of the i-th air conditioner; According to the basic power and the dynamic power To obtain the charging and discharging power The expression is: ; Obtain the power status of a single air conditioner. The state of electrical energy The expression is: ； Wherein, the C i Let the heat capacity of the i-th air conditioner be denoted as . The deformation thermal dynamic equilibrium equation and the basic power The expression and the state of electrical energy By combining the expressions, we get Wherein α i =1 / C i R i , where is the self-discharge coefficient of the i-th air conditioner; After discretization, the expression for the preliminary battery model is obtained. Wherein α i =1-Δt / C i R i The Let k be the electrical energy state of the i-th air conditioner at time k. Let k be the charging and discharging power of the i-th air conditioner at time k. Let Δt be the electrical energy state of the i-th air conditioner at time k+1, and let Δt be the time change between time k and time k+1. The basic power P of the air conditioner aggregator is set to maintain the set temperature at time k. b (k), to obtain the expression ; The formula for calculating the total power of an air conditioning aggregator is as follows: ； The upper and lower boundary values ​​of the total power of the air conditioning aggregator were calculated. and Subtract the basic power P of the air conditioner aggregator. b (k), to obtain the Upper and lower boundary values: ； The formulas for obtaining the conditions for minimizing and maximizing the expected temperature deviation of the i-th air conditioner are as follows: ； The upper and lower boundary values ​​of the aggregation power of the air conditioner aggregator are calculated according to the aforementioned conditional formula. and Combined with the expression of the preliminary battery model , to obtain the Upper and lower boundary values: ； Wherein, the self-discharge coefficient Let K be the average self-discharge coefficient of n air conditioners, where K is a positive integer greater than or equal to k; Based on the expression of the preliminary battery model, the upper and lower boundary values ​​of the charge and discharge power, and the upper and lower boundary values ​​of the state of energy, the expression of the general battery model for air conditioner aggregators is obtained: 。 6. The air conditioning load prediction model establishment system according to claim 4, characterized in that, The uncertainty calculation module is also used to perform: Define the user's willingness to respond to the user's air conditioning electricity consumption habits for the i-th air conditioner as... The user's willingness to respond The expression is: ； Wherein, the ρ i It is user sensitivity, the aforementioned As the incentive level, the λ i Regarding the impact of weather factors on users, the Where k is a positive real number, the The difference between indoor temperature and standard comfort temperature is λ0, which represents the user's willingness to respond when the difference between indoor temperature and standard comfort temperature is at its maximum. Obtain the user's response intention The membership function expression is: ； Among them, the The upper limit of the user's willingness to respond, the As the lower limit of the user's willingness to respond, the The average value of user response intention, the The above and the aforementioned The expression is: ； The user's response intention Represented as a fuzzy random function: ； The ρ i Satisfying a normal distribution, the and stated respectively ρ i The mean and standard deviation, the As a fuzzy variable, the It is a random variable.

7. A method for predicting air conditioning load, characterized in that, include: Collect the daytime temperature data, predict the daytime temperature data based on the daytime temperature data, and predict the predicted temperature value for the next time based on the daytime temperature data. Based on the previous day's temperature data, the intraday temperature data, the predicted temperature value, and the air conditioning load prediction model considering uncertainties generated in the air conditioning load prediction model establishment method according to claims 1-3, the air conditioning load at the next moment is predicted.

Images