Campus energy system heating power data-driven control method considering temperature change

Abstract

A data-driven control method for heating power in a park energy system, taking into account temperature variations, aims to solve the power control problem of heating equipment in such systems. It fully considers user thermal comfort needs and online system scheduling requirements, establishing a model-free adaptive prediction model for heating power. This model is then modified based on indoor temperature measurements to obtain a data-driven model for heating power based on indoor temperature measurements. The controlled power of the system's heat source equipment is obtained by solving an optimization problem. Using this data-driven control method for heating power in a park energy system, online control of the system's heat source equipment power can be achieved, effectively reducing indoor temperature fluctuations and economically and reliably meeting the system's energy needs. This method effectively solves the heating power control problem and is of great significance for the scheduling and operation of park energy systems.

Description

Technical Field

[0001] This invention relates to a data-driven control method for heating power in a park energy system. In particular, it relates to a data-driven control method for heating power in a park energy system that takes into account temperature changes. Background Technology

[0002] With the continuous growth of energy demand and carbon emissions, carbon neutrality has become a common goal for countries around the world. Developing and utilizing renewable energy and improving energy efficiency are inevitable choices for achieving carbon reduction and sustainable development. The park's energy system integrates electricity, natural gas, cooling, and heating systems, with these different energy systems closely coupled. Through the coordinated optimization and complementary operation of multiple energy sources in production, conversion, and consumption, it not only meets the flexible supply of electricity, cooling, and heating to users but also significantly improves energy efficiency.

[0003] The park's energy system is a complex multi-energy coupled system. Electricity demand can be met by the external power grid, photovoltaic power, wind turbines, and combined heat and power (CHP) equipment, offering high reliability and strong regulation capabilities. Summer cooling and winter heating are provided by ground source heat pumps, compressor refrigeration equipment, gas boilers, and electric boilers. Their power output, along with the external environment, determines the indoor temperature variation characteristics, thus affecting user thermal comfort. Simultaneously, changes in outdoor temperature, fluctuations in sunlight characteristics, indoor occupant activity, and changes in building heat dissipation parameters all affect the building's thermal characteristics. Therefore, dynamic correction of the heating load is necessary to adjust the heating power of the heat source equipment to economically and reliably meet user energy needs. Due to the significant uncertainty caused by the superposition of numerous influencing factors, the impact of changes in building thermal characteristics on user heat load must be fully considered during the optimization and scheduling process.

[0004] Currently, the optimization of energy scheduling in industrial parks to meet user thermal comfort and address heating load fluctuations often relies on establishing advance scheduling models that consider the uncertainties of external environmental changes. However, these models fail to address the real-time adjustment needs during actual system operation. Alternatively, rule-based control of heating power enables real-time system adjustments but cannot meet the scheduling requirements for optimized system operation. Therefore, there is an urgent need for a heating power control method that can meet user thermal comfort while considering system operation scheduling needs. This method should be based on feedback control theory and data-driven scheduling principles, coordinating user energy demand and the operation control of heat source devices based on indoor temperature measurement data to economically and reliably meet the energy needs of the industrial park's energy system. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a data-driven control method for heating power of a park energy system that takes into account temperature changes and enables online control of the power of the system's heat source equipment, effectively reducing indoor temperature fluctuations.

[0006] The technical solution adopted in this invention is: a data-driven control method for heating power of a park energy system that takes into account temperature changes, comprising the following steps:

[0007] 1) Based on the selected park energy system, input the system heating equipment composition and operating parameters; input the park building thermal parameters, including the building heat dissipation coefficient a, the building heat dissipation equivalent area F, and the building solar radiation equivalent area S; input the indoor temperature measurement interval Δt; set the heating power scheduling control step size ΔT = n·Δt, where n is the total number of measurements for each measurement target within a control period; set the total number of heating power control periods N; and set the indoor temperature target value T. ref Set the weighting coefficient μ for the change in the pseudo-Jacobi matrix and the weighting coefficient λ for the change in heating power; initialize the control period k=1, and initialize the indoor temperature, outdoor temperature, solar irradiance, and power of the heat source equipment based on the current measurement values;

[0008] 2) At the end time t of the kth control period, read n indoor temperature measurements and power command values ​​of heat source equipment during the kth control period, read n outdoor temperature measurements and n solar irradiance measurements during the kth control period, and calculate the equivalent heat load power X(k) of the system during the kth control period; input m predicted outdoor temperature data and m predicted solar irradiance data for the k+1 control period, where m is the total number of predictions for each prediction target within a control period. To ensure the simultaneity of the measured data and predicted data within the control period, m = n is taken.

[0009] 3) Establish a data-driven heating power feedback control model, and modify the data-driven heating power feedback control model based on the indoor temperature status to obtain the modified heating power control model of the park's energy system.

[0010] 4) Based on the park energy system heating power control correction model established in step 3) and the indoor temperature target value set in step 1), take the minimum weighted sum of indoor temperature deviation and heating power change as the objective function, consider the system heating equipment operation constraints, solve the linear programming problem, and obtain the system heating power command and the corresponding expected indoor temperature of the system in the k+1 control period.

[0011] 5) Output results, including: the expected indoor temperature of the system in the (k+1)th control period, and the system heating power command in the (k+1)th control period, and send them to the heating equipment for execution;

[0012] 6) Update the scheduling control period to k = k + 1, and determine whether k is greater than the total number of heating power control periods N. If it is greater, end the process; otherwise, return to step 2.

[0013] This invention presents a data-driven control method for heating power in a park energy system that considers temperature variations. Based on solving the power control problem of heating equipment in park energy systems, it fully considers user thermal comfort needs and online system scheduling requirements. A model-free adaptive prediction model for heating power is established and then modified based on the model-free adaptive prediction model obtained from indoor temperature measurements, resulting in a data-driven model for heating power based on indoor temperature measurements. The controlled power of the system's heat source equipment is obtained by solving an optimization problem. Using this data-driven control method for heating power in a park energy system, online control of the system's heat source equipment power can be achieved, effectively reducing indoor temperature fluctuations and economically and reliably meeting the system's energy needs. This method effectively solves the heating power control problem and is of great significance for the scheduling and operation of park energy systems.

[0014] The present invention provides a data-driven control method for heating power in a park energy system that takes into account temperature changes. Its theoretical basis is the use of a newly introduced concept of pseudo-partial derivatives or pseudo-Jacobi matrices to replace the general nonlinear system with a dynamic linear time-varying model near the controlled system trajectory. This method does not require knowledge of the detailed mathematical model of the park energy system; it only needs to establish a data model based on real-time operating data to dynamically and adaptively characterize the system's input-output relationship, thereby achieving the goal of controlling the heating power of the park energy system. Compared to traditional heat source power control methods, the method of the present invention can fully utilize the measured indoor temperature as feedback to adjust the heating power in a timely manner, achieving better indoor heating effects and accurately controlling the difference between the indoor temperature and the target heating value. Furthermore, in actual operation, accurate parameters of the park heat source system are difficult to obtain due to the influence of the park energy system's operating conditions and the external environment; in addition, due to the strong time-varying nature and large time constant of its thermodynamic characteristics, indoor heat load demand exhibits significant randomness and fluctuation. Therefore, compared with detailed solution methods based on refined modeling, the method of this invention does not rely on a large number of accurate system parameters. It can not only achieve the goal of controlling the system's heat source power, but also avoid control deviations caused by inaccurate parameters in the system. Attached Figure Description

[0015] Figure 1 This is a flowchart of the data-driven control method for heating power of a park energy system that takes into account temperature changes, according to the present invention.

[0016] Figure 2 These are outdoor temperature curves and solar irradiance curves;

[0017] Figure 3 This is the indoor temperature curve for a scenario without heating;

[0018] Figure 4 It is an indoor temperature curve obtained by using constant power control and the control method of this patent;

[0019] Figure 5 It is a heating power curve obtained by using constant power control and the control method of this patent. Detailed Implementation

[0020] The following detailed description of the heating power data-driven control method for a park energy system that takes into account temperature changes, in conjunction with embodiments and accompanying drawings, provides a more comprehensive understanding of the present invention.

[0021] like Figure 1 As shown, the heating power data-driven control method for a park energy system that takes into account temperature changes according to the present invention includes the following steps:

[0022] 1) Based on the selected park energy system, input the system heating equipment composition and operating parameters; input the park building thermal parameters, including the building heat dissipation coefficient a, the building heat dissipation equivalent area F, and the building solar radiation equivalent area S; input the indoor temperature measurement interval Δt; set the heating power scheduling control step size ΔT = n·Δt, where n is the total number of measurements for each measurement target within a control period; set the total number of heating power control periods N; and set the indoor temperature target value T. ref Set the weighting coefficient μ for the change in the pseudo-Jacobi matrix and the weighting coefficient λ for the change in heating power; initialize the control period k=1, and initialize the indoor temperature, outdoor temperature, solar irradiance, and power of the heat source equipment based on the current measurement values;

[0023] 2) At the end time t of the kth control period, read n indoor temperature measurements and power command values ​​of heat source equipment during the kth control period, read n outdoor temperature measurements and n solar irradiance measurements during the kth control period, and calculate the equivalent heat load power X(k) of the system during the kth control period; input m predicted outdoor temperature data and m predicted solar irradiance data for the k+1 control period, where m is the total number of predictions for each prediction target within a control period. To ensure the simultaneity of the measured data and predicted data within the control period, m = n is taken.

[0024] The calculated equivalent heat load power X(k) of the system during the k-th control period is expressed as follows:

[0025] X(k)=[x(tn·Δt),…,x(t-2·Δt),x(t-Δt)] T (1)

[0026] x(tl·Δt)=P s (k)-P out (tl·Δt),l=1,2,…,n (2)

[0027] In the formula, X(k) represents the system equivalent heat load power during the k-th control period, x(tn·Δt) represents the system equivalent heat load power at time tn·Δt, x(t-Δt) represents the system equivalent heat load power at time t-Δt, n is the total number of measurements for each measurement target within a control period, time t is the end time of the k-th control period, and time tn·Δt is the start time of the k-th control period; x(tl·Δt) represents the system equivalent heat load power at time tl·Δt, and l is the l-th measurement for each measurement target within a control period; P s (k) represents the system heating power during the k-th control period, which is obtained from historical control command values; P out (tl·Δt) represents the equivalent power of heat dissipation from the building to the outside at time tl·Δt, and its expression is:

[0028] P out (tl·Δt)=aF(T(tl·Δt)-T out (tl·Δt))-SE sun (tl·Δt),l=1,2,…,n (3)

[0029] In the formula, a represents the building heat dissipation coefficient, F represents the building's equivalent heat dissipation area, S represents the building's equivalent solar radiation receiving area; T(tl·Δt) represents the measured indoor temperature at time tl·Δt; T out (tl·Δt) represents the measured outdoor temperature at time tl·Δt; E sun (tl·Δt) represents the measured value of solar irradiance at time tl·Δt.

[0030] 3) Establish a data-driven heating power feedback control model, and modify the data-driven heating power feedback control model based on the indoor temperature status to obtain the modified heating power control model of the park's energy system.

[0031] The data-driven heating power feedback control model is specifically represented as follows:

[0032]

[0033] In the formula, This represents the predicted indoor temperature value estimated by the controller for the (k+1)th control period; T(t) represents the measured indoor temperature value at time t, where time t is the end time of the kth control period; I is a unit column vector, represented as I = [1, 1, ..., 1]. T ; Φ(k) represents the pseudo-Jacobi matrix of the change in indoor temperature with respect to the equivalent heat load power of the system during the k-th control period; ΔX(k+1) represents the change in the equivalent heat load power of the system during the (k+1)-th control period; And ΔX(k+1) are expressed as:

[0034]

[0035]

[0036]

[0037] ΔX(k+1)=[Δx(t), Δx(t+Δt),…, Δx(t+(n-1)·Δt)] (8)

[0038] Δx(t+l·Δt)=x(t+(l+1)·Δt)-x(t+l·Δt),l=0,1,…,n-1 (9)

[0039] In the formula, This represents the predicted indoor temperature at time t+n·Δt, estimated by the controller. Δx(t) represents the predicted indoor temperature at time t+Δt estimated by the controller, n is the total number of measurements for each measurement target within a control period, and time t+n·Δt is the end time of the (k+1)th control period; Δx(t) represents the change in the system's equivalent heat load power at time t, Δx(t+(n-1)·Δt) represents the change in the system's equivalent heat load power at time t+(n-1)·Δt; Δx(t+l·Δt) represents the change in the system's equivalent heat load power at time t+l·Δt, l is the lth measurement for each measurement target within a control period; x(t+(l+1)·Δt) represents the system's equivalent heat load power at time t+(l+1)·Δt, x(t+l·Δt) represents the system's equivalent heat load power at time t+l·Δt.

[0040] In the data-driven heating power feedback control model, the method for estimating the pseudo-Jacobi matrix Φ(k) of the change in equivalent heat load power of the system due to indoor temperature is specifically expressed as follows:

[0041]

[0042]

[0043] ΔX(k)=[Δx(tn·Δt),…,Δc(t-2·Δt),Δx(t-Δt)] T (12)

[0044] Δx(tl·Δt)=x(t-(l-1)·Δt)-x(tl·Δt),l=1,2,…,n (13)

[0045] In the formula, Φ(k) and Φ(k-1) represent the pseudo-Jacobi matrices of the change in indoor temperature with respect to the equivalent heat load power of the system during control periods k and k-1, respectively; T(k) represents the measured value of indoor temperature during the k-th control period. This represents the predicted indoor temperature value estimated by the controller for the k-th control period; μ represents the weighting coefficient of the change in the pseudo-Jacobi matrix; T(tn·Δt) represents the measured indoor temperature value at time tn·Δt, where n is the total number of measurements for each measurement target within a control period, time t is the end time of the k-th control period, and time tn·Δt is the start time of the k-th control period; I is a unit column vector, represented as I = [1, 1, ..., 1]. T ΔX(k) represents the change in the system's equivalent heat load power during the k-th control period, Δx(tn·Δt) represents the change in the system's equivalent heat load power at time tn·Δt, and Δx(t-Δt) represents the change in the system's equivalent heat load power at time t-Δt; Δx(tl·Δt) represents the change in the system's equivalent heat load power at time tl·Δt, where l is the l-th measurement of each measurement target within a control period; x(t-(l-1)·Δt) represents the system's equivalent heat load power at time t-(l-1)·Δt, and x(tl·Δt) represents the system's equivalent heat load power at time tl·Δt; T(k) is specifically represented as:

[0046] T(k)=[T(t-(n-1)·Δt),…,T(t-Δt),T(t)] T (14)

[0047] In the formula, T(t-(n-1)·Δt) represents the indoor temperature measurement at time t-(n-1)·Δt, and T(t) represents the indoor temperature measurement at time t.

[0048] The calculation of Φ(k) requires the value of Φ(k-1) as a basis. Based on the recursive relationship, the value of Φ(1) needs to be obtained. Φ(1) is initialized based on the sensitivity of the system to changes in the equivalent heat load power of the indoor temperature, specifically expressed as:

[0049]

[0050] In the formula, φ(0) represents the partial derivative of the indoor temperature at the initial time t=0 with respect to the change in the equivalent heat load power of the system; φ(Δt) represents the partial derivative of the indoor temperature at time Δt with respect to the change in the equivalent heat load power of the system; φ(l·Δt) represents the partial derivative of the indoor temperature at time l·Δt with respect to the change in the equivalent heat load power of the system; and φ((n-1)·Δt) represents the partial derivative of the indoor temperature at time (n-1)·Δt with respect to the change in the equivalent heat load power of the system.

[0051] The partial derivative of the change in indoor temperature with respect to the change in the system's equivalent heat load power at different times can be calculated based on the measured indoor temperature and the change in the system's equivalent heat load power, expressed as:

[0052] T(Δt)=T(0)+φ(0)·Δx(0) (16)

[0053]

[0054] In the formula, T(0) represents the initial value of the indoor temperature at the initial t=0, T(Δt) represents the measured value of the indoor temperature at the time Δt, T(l·Δt) represents the measured value of the indoor temperature at the time l·Δt, T((l-1)·Δt) represents the measured value of the indoor temperature at the time (l-1)·Δt, and T((l-2)·Δt) represents the measured value of the indoor temperature at the time (l-2)·Δt; Δx(0) represents the change in the equivalent heat load power of the system at the initial t=0, Δx((l-1)·Δt) represents the change in the equivalent heat load power of the system at the time (l-1)·Δt, and Δx((l-2)·Δt) represents the change in the equivalent heat load power of the system at the time (l-2)·Δt.

[0055] The aforementioned data-driven heating power feedback control model based on indoor temperature status is modified to obtain a modified heating power control model for the park's energy system, specifically expressed as follows:

[0056] Φ e (k)=[e(t)·IE(k)-ΔT ref (k)] / ΔX(k) (18)

[0057] Φ * (k)=Φ(k)-Φ e (k) (19)

[0058]

[0059] In the formula, e(t) represents the difference between the measured indoor temperature at time t and the target value, where time t is the end time of the k-th control period; I is a unit column vector, represented as I = [1, 1, ..., 1]. T E(k) represents the difference between the measured indoor temperature and the target value during the k-th control period; Φ e (k) represents the pseudo-Jacobi matrix of the change in equivalent heat load power of the system due to the indoor temperature deviation during the k-th control period; Δk(k) represents the change in equivalent heat load power of the system during the k-th control period; ΔT ref (k) represents the change in the target indoor temperature during the k-th control period; Φ *(k) represents the pseudo-Jacobi matrix of the change in indoor temperature with respect to the equivalent heat load power of the system during the k-th control period after correction; This represents the system's expected indoor temperature for the (k+1)th control period after correction; E(k) and ΔT ref (k) is specifically represented as:

[0060] E(k)=[e(t-Δt), e(t-2·Δt),…, e(tn·Δt)] T (twenty one)

[0061] e(tl·Δt)=T(tl·Δt)-T ref (tl·Δt),l=1,2,…,n (22)

[0062] ΔT ref (k)=[ΔT ref (t-Δt),ΔT ref (t-2·Δt),…,ΔT ref (tn·Δt)] T (twenty three)

[0063] ΔT ref (tl·Δt)=T ref (t-(l-1)·Δt)-T ref (tl·Δt),l=1,2,…,n (24)

[0064] In the formula, e(t-Δt) represents the difference between the measured indoor temperature value and the target value at time t-Δt; e(tn·Δt) represents the difference between the measured indoor temperature value and the target value at time tn·Δt; e(tl·Δt) represents the difference between the measured indoor temperature value and the target value at time tl·Δt, where l is the l-th measurement of each measurement target within a control period; ΔT ref (t-Δt) represents the change in the target indoor temperature at time t-Δt, where ΔT ref (tn·Δt) represents the change in the target indoor temperature at time tn·Δt, where n is the total number of measurements for each target within a control period, and time tn·Δt is the start time of the k-th control period; ΔT ref (tl·Δt) represents the change in the target indoor temperature at time tl·Δt, where T(tl·Δt) and T ref (tl·Δt) represent the measured indoor temperature and the target indoor temperature at time tl·Δt, respectively.

[0065] 4) Based on the park energy system heating power control correction model established in step 3) and the indoor temperature target value set in step 1), take the minimum weighted sum of indoor temperature deviation and heating power change as the objective function, consider the system heating equipment operation constraints, solve the linear programming problem, and obtain the system heating power command and the corresponding expected indoor temperature of the system in the k+1 control period.

[0066] The objective function, which is to minimize the weighted sum of indoor temperature deviation and heating power variation, is specifically expressed as follows:

[0067]

[0068] In the formula, P s (k+1), P s (k) represents the system heating power command value for the (k+1)th and kth control periods, respectively; λ represents the weighting coefficient of the heating power change. T represents the system's expected indoor temperature during the (k+1)th control period after correction. ref (k+1) represents the target indoor temperature value for the (k+1)th control period;

[0069] The operating constraints of the system heating equipment are expressed as follows:

[0070]

[0071]

[0072] In the formula, Indicates the upper limit of the heating power of the heat source equipment; This indicates the upper limit of the change in heating power of a heat source device within a control interval.

[0073] 5) Output results, including: the expected indoor temperature of the system in the (k+1)th control period, and the system heating power command in the (k+1)th control period, and send them to the heating equipment for execution;

[0074] 6) Update the scheduling control period to k = k + 1, and determine whether k is greater than the total number of heating power control periods N. If it is greater, end the process; otherwise, return to step 2.

[0075] The following are specific examples:

[0076] Based on the selected park energy system, the input heat source equipment is an electric boiler with an energy efficiency of 0.95. The maximum capacity of the electric boiler system is 1500kW, and the minimum output is 0kW. Inputs include changes in outdoor temperature and outdoor irradiance, such as... Figure 2 , Figure 3 As shown. The heating power dispatch control interval is set to ΔT = 15 min, and the target indoor temperature T controlled by the heat source equipment is set.ref =20℃. Number of measurements n = 5 within one scheduling cycle.

[0077] To verify the effectiveness of the method proposed in this invention, the following three schemes were used to control and compare the heating power of the park's energy system.

[0078] Option 1: The heat source equipment does not operate;

[0079] Option 2: The output power of the heat source equipment remains constant;

[0080] Option 3: Adopt a control method driven by heating power data of the park's energy system that takes into account temperature changes;

[0081] The computer hardware environment for performing the optimized calculations was an Intel(R) Core(TM) i7-9700 CPU with a clock speed of 3.00GHz and 24GB of memory; the software environment was a Windows 10 operating system.

[0082] Table 1 Comparison of Indoor Temperature Results in Different Scenarios

[0083]

[0084] Table 1 shows a comparison of indoor temperature changes under the three control schemes. Figure 3 , Figure 4 A comparison of indoor temperatures reveals that while the output of heat source equipment can increase indoor temperature, constant power control leads to significant temperature fluctuations, affecting user thermal comfort. In contrast, a data-driven control method for the park's energy system that considers temperature changes ensures that the heat source equipment's output power is controlled accordingly. Figure 5 As shown, the output of the heat source equipment can be continuously adjusted according to the indoor temperature, resulting in better indoor temperature control and a more closely aligned response to user energy needs. Under the same energy consumption conditions, a stable energy supply for the system is achieved. In summary, the temperature-dependent data-driven control method for heating power in park energy systems proposed in this invention can effectively solve the heating power control problem and is of great significance for the scheduling and operation of park energy systems.

Claims

1. A data-driven control method for heating power in a park energy system that takes into account temperature changes, characterized in that, Includes the following steps: 1) Based on the selected park energy system, input the system heating equipment composition and equipment operating parameters, and input the park building thermal parameters, including the building heat dissipation coefficient. Building heat dissipation equivalent area The equivalent area of ​​solar radiation received by a building Input indoor temperature measurement interval Set the heating power scheduling control step size ,in To determine the total number of measurements for each measurement target within a control period, set the total number of control periods for heating power. Set indoor temperature target value Set the weighting coefficients for the changes in the pseudo-Jacobi matrix. Weighting coefficient for changes in heating power Initialize control period The indoor temperature, outdoor temperature, solar irradiance, and power of the heat source equipment are initialized based on the current measurement values. 2) in the End time of each control period Read the first Indoor temperature during each control period Read the first measurement value, the power command value of the heat source equipment, and the first... Outdoor temperature during each control period Individual measurements, solar irradiance The nth measurement value is calculated to obtain the nth measurement value. The system equivalent heat load power during each control period ;enter Outdoor temperature during control period Predicted data, solar irradiance One predicted data, of which This represents the total number of predictions for each target within a control period. To ensure the simultaneity of measurement and prediction data within the control period, we take... ; 3) Establish a data-driven heating power feedback control model, and modify the data-driven heating power feedback control model based on the indoor temperature status to obtain the modified heating power control model of the park's energy system. 4) Based on the heating power control correction model of the park energy system established in step 3) and the indoor temperature target value set in step 1), with the weighted sum of indoor temperature deviation and heating power variation as the objective function, considering the operating constraints of the system heating equipment, solve the linear programming problem to obtain the ... The system heating power command and corresponding number of control periods The expected indoor temperature of the system during the control period; 5) Output results, including: the first... The system's expected indoor temperature during the control period, the first The system heating power command for the control period is issued and sent to the heating equipment for execution. 6) Update the scheduling control period to ,judge Is it greater than the total number of heating power control periods? If it is greater than 0, then the process ends; otherwise, return to step 2.

2. The data-driven control method for heating power of a park energy system taking into account temperature changes, as described in claim 1, is characterized in that... Step 2) The calculation yields the first The system equivalent heat load power during each control period The expression is as follows: ； ； In the formula, Indicates the first The system's equivalent heat load power during each control period express The system's equivalent heat load power at any given time, express The system's equivalent heat load power at any given time, The total number of measurements for each measurement target within a control period, at time... For the first The end time of each control period, time For the first The start time of each control period; express The system equivalent heat load power at time t, l is the lth measurement of each measurement target within a control period; Indicates the first The system heating power for each control period is obtained from historical control command values; Indicates in The equivalent power of heat dissipation from the building to the outside at any given time is expressed as: ； In the formula, Indicates the building's heat dissipation coefficient. This represents the equivalent area for heat dissipation in a building. This represents the equivalent area of ​​solar radiation received by the building; express The measured indoor temperature at any given time; express Outdoor temperature measurement at any time; express The measured value of solar irradiance at a given time.

3. The data-driven control method for heating power of a park energy system considering temperature changes, as described in claim 1, is characterized in that... The data-driven heating power feedback control model mentioned in step 3) is specifically represented as follows: ； In the formula, Indicates the first estimated value by the controller Predicted indoor temperature during the control period; express Indoor temperature measurement at time , time For the first The end time of each control period; For a unit column vector, it is represented as ; Indicates the first The pseudo-Jacobi matrix of indoor temperature during the control period on the change of equivalent heat load power of the system; Indicates the first The change in the system's equivalent heat load power during the control period; and They are represented as follows: ； ； ； In the formula, Indicates the controller's estimate Predicted indoor temperature at any given time. Indicates the controller's estimate Predicted indoor temperature at any given time. The total number of measurements for each measurement target within a control period, at time... For the first The end time of the control period; express The change in the system's equivalent heat load power at any given time. express The change in the system's equivalent heat load power at any given time; express The change in the system's equivalent heat load power at a given time, l is the l-th measurement of each measurement target within a control period; express The system's equivalent heat load power at any given time, express The system's equivalent heat load power at any given time.

4. The data-driven control method for heating power of a park energy system taking into account temperature changes, as described in claim 3, is characterized in that... In the data-driven heating power feedback control model, the pseudo-Jacobi matrix of indoor temperature on the change of the system's equivalent heat load power is... The estimation method, Specifically, it is expressed as follows: ； ； ； ； In the formula, They represent The pseudo-Jacobi matrix of indoor temperature during the control period on the change of equivalent heat load power of the system; Indicates the first Indoor temperature measurements during each control period Indicates the first estimated value by the controller Predicted indoor temperature for each control period; Weighting coefficients representing the changes in the pseudo-Jacobi matrix; express Indoor temperature measurements at various times, among which The total number of measurements for each measurement target within a control period, at time... For the first The end time of each control period, time For the first The start time of each control period; For a unit column vector, it is represented as ; Indicates the first Changes in the system's equivalent heat load power during each control period express The change in the system's equivalent heat load power at any given time. express The change in the system's equivalent heat load power at any given time; express The change in the system's equivalent heat load power at a given time, l is the l-th measurement of each measurement target within a control period; express The system's equivalent heat load power at any given time, express The system's equivalent heat load power at any given moment; Specifically, it is expressed as follows: ； In the formula, express The indoor temperature measurement at that moment. express The measured indoor temperature at any given time; The calculation requires Based on the value of , and following the recursive relation, we need to find . The value, The system's sensitivity to changes in equivalent heat load power based on indoor temperature is initialized, specifically as follows: ； In the formula, Indicates the initial The partial derivative of the indoor temperature at any given time with respect to the change in the system's equivalent heat load power; express The partial derivative of the indoor temperature at any given time with respect to the change in the system's equivalent heat load power; express The partial derivative of the indoor temperature at time t with respect to the change in the system's equivalent heat load power. express The partial derivative of the indoor temperature at any given time with respect to the change in the system's equivalent heat load power; The partial derivative of the change in indoor temperature with respect to the change in the system's equivalent heat load power at different times can be calculated based on the measured indoor temperature and the change in the system's equivalent heat load power, expressed as: ； ； In the formula, Indicates the initial The initial indoor temperature at that moment. express The indoor temperature measurement at that moment. express The indoor temperature measurement at that moment. express The indoor temperature measurement at that moment. express The measured indoor temperature at any given time; Indicates the initial The change in the equivalent heat load power of the system at any given time. express The change in the equivalent heat load power of the system at any given time. express The change in the equivalent heat load power of the system at any given time.

5. The data-driven control method for heating power of a park energy system taking into account temperature changes, as described in claim 1, is characterized in that... Step 3) involves modifying the data-driven heating power feedback control model based on indoor temperature conditions to obtain the modified heating power control model for the park's energy system, specifically expressed as follows: ； ； ； In the formula, express The difference between the measured indoor temperature at time 1 and the target value, where time 1 is the indoor temperature at time 2. For the first The end time of each control period; For a unit column vector, it is represented as ; Indicates the first The difference between the measured indoor temperature value and the target value during each control period; Indicates the first The pseudo-Jacobi matrix of the effect of indoor temperature deviation on the change of equivalent heat load power of the system during each control period; Indicates the first The change in the system's equivalent heat load power during each control period; Indicates the first Changes in the target indoor temperature during each control period; Indicates the first The pseudo-Jacobi matrix of indoor temperature during the control period on the change of equivalent heat load power of the system; Indicates the corrected number The pseudo-Jacobi matrix of indoor temperature during the control period on the change of equivalent heat load power of the system; Indicates the corrected number The expected indoor temperature of the system during the control period; and Specifically, it is expressed as follows: ； ； ； ； In the formula, express The difference between the measured indoor temperature at any given time and the target value. express The difference between the measured indoor temperature and the target value at any given time; express The difference between the measured indoor temperature value and the target value at a given time, where l is the l-th measurement of each measurement target within a control period; express The change in the target indoor temperature at any given time. express The change in the target indoor temperature at any given time, where The total number of measurements for each measurement target within a control period, at time... For the first The start time of each control period; express The change in the target indoor temperature at any given time. They represent The measured indoor temperature value at any given time and the target indoor temperature value.

6. The data-driven control method for heating power of a park energy system taking into account temperature changes, as described in claim 1, is characterized in that... Step 4), which uses the weighted sum of indoor temperature deviation and heating power variation as the objective function, is specifically expressed as follows: ； In the formula, They represent the first The system heating power command value during the control period; Weighting coefficients representing changes in heating power; Indicates the corrected number The expected indoor temperature of the system during the control period; Indicates the first The target value for indoor temperature during the control period.

Images