A heating load correction method based on indoor temperature

Abstract

The application discloses a heat supply load correction method based on indoor temperature, and mainly comprises the following steps: step S1, obtaining historical operation data of the heat station at each time and outdoor meteorological parameters; step S2, collecting building indoor temperature T in,0 at a certain time, calculating building heat exchange characteristic parameter BC0; step S3, calculating actual indoor temperature at all times; step S4, setting indoor temperature T in,set,i at each time, calculating the sum T sum2,set,i of the set supply and return water temperatures of the two networks that should be reached at each time; step S5, establishing a heat exchanger support vector regression (SVR) black box model, obtaining operation parameter setting m 1,set,i of the heat network required under the set indoor temperature T in,set,i ; step S6, according to a heat exchanger white box physical model, solving the corrected operation parameters of the first network and the second network under the condition of setting the first network flow m 1,set,i , and calculating the corrected first network load Q 1,correct,i .

Description

Technical Field

[0001] This invention relates to the field of heating pipe networks, and more specifically to a method for correcting heating load based on indoor temperature. Background Technology

[0002] In my country, building energy consumption accounts for approximately 30% of the country's total energy consumption, and HVAC systems account for about 50% of this. HVAC equipment capacity is designed for extreme conditions; if accurate load information is lacking, supplying too much heat to the building will result in significant energy waste, while supplying too little heat will lead to decreased indoor thermal comfort, causing resident dissatisfaction and complaints. Therefore, accurately grasping changes in building heat load not only provides a basis for formulating energy system operation strategies but also helps avoid unnecessary energy waste while ensuring indoor thermal comfort.

[0003] Currently, heat load prediction is mainly divided into three categories: white-box method, black-box method, and gray-box method. The white-box method is based on detailed physical equations and requires a large amount of building information for heat load calculation. However, in actual engineering projects, sufficient physical parameter information is often unavailable to meet the requirements of load calculation. Therefore, the black-box method, based on historical data, is more feasible. However, existing black-box methods use historical data as a reference without considering the possibility of excessive or insufficient heat supply in those historical data. If heat load prediction is based solely on such historical data, future data will still show instances of excessive or insufficient heat supply. Therefore, it is necessary and reasonable to correct the actual heat load operation data based on indoor temperature feedback, and then use the corrected data for black-box prediction. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art by providing a heating load correction method based on indoor temperature.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A heating load correction method based on indoor temperature mainly includes the following steps:

[0007] Step S1: Obtain the historical operating data of the heating station at various times and its corresponding outdoor meteorological parameters. The historical operating data includes the primary water supply temperature T. s1,i (°C), return water temperature of the first network T r1,i (°C), network traffic m 1,i (kg / s), secondary water supply temperature T s2,i (°C), Secondary network return water temperature T r2,i (°C), Secondary network traffic m 2,i(kg / s), outdoor meteorological parameters include outdoor dry-bulb temperature T out,i (°C), where the subscript i indicates time i.

[0008] Step S2: Collect the building's indoor temperature at a certain moment, denoted as T. in,0 (°C), and based on the current outdoor dry-bulb temperature T. out,0 Secondary water supply temperature T s2,0 Secondary network return water temperature T r2,0 According to the formula The heat exchange characteristic parameter BC0 of the building was calculated.

[0009] Step S3: Based on the secondary network water supply temperature T at each time point s2,i Secondary network return water temperature T r2,i and outdoor dry-bulb temperature T out,i And the building heat transfer characteristic parameter BC0 obtained in step S2, to calculate the actual indoor temperature at all times.

[0010] Step S4: Set the indoor temperature at each time point to T according to actual needs. in,set,i According to formula T sum2,set,i =2BC0(T in,set,i -T out,i )+T in,set,i (3) The sum of the set supply and return water temperatures of the secondary network that should be reached at each time under the condition of meeting the set indoor temperature was calculated, T. sum2,set,i .

[0011] Step S5: Establish a support vector regression (SVR) black box model of the heat exchanger to obtain the set indoor temperature T. in,set,i The required operating parameters for the heating network are set as follows: Network flow rate (m³) 1,set,i .

[0012] Step S501: Based on the actual operating data of the heating station at each time, a support vector regression (SVR) model for the heat exchanger is trained. The input parameters are the primary network supply water temperature, secondary network flow rate, secondary network supply water temperature and secondary network return water temperature at each time. The output parameter is the primary network flow rate at the corresponding time.

[0013] Step S502: Maintain the set temperature difference of the secondary network at each moment to be the same as the actual operating data, i.e., T s2,set,i -T r2,set,i =T s2,i -T r2,i The sum of the supply and return water temperatures of the secondary network at each time point obtained according to step S4 is T. sum2,set,i It can solve for the set water supply temperature T of the secondary network at each moment. s2,set,i The set return water temperature T of the secondary network r2,set,i .

[0014] Step S503: Calculate the actual water supply temperature T at each time point. s1,i Actual secondary network traffic m 2,i Set the secondary water supply temperature T s2,set,i Set the secondary network return water temperature T r2,set,i Using the support vector regression (SVR) model of the heat exchanger as input data, the set primary flow rate m at each time step is obtained. 1,set,i .

[0015] Step S6: Based on the physical model of the heat exchanger white box, and according to the set primary flow rate m... 1,set,i Solve for the corrected network load Q 1,set,i .

[0016] Step S601: Establish the physical model of the heat exchanger according to the following formula.

[0017] Q 1,i =cm 1.i (T s1,i -T r1,i (4)

[0018] Q 2.i =cm 2.i (T s2,i -T r2,i (5)

[0019] η i ×Q 1,i =Q 2.i (6)

[0020]

[0021] k i =am 1.i +b (8)

[0022] In the formula, Q 1,i For heat exchange in a single network, W; Q 2.i For heat exchange in the secondary network, W; η i For heat exchange efficiency, k i The heat transfer coefficient is W / (m²). 2 ·K); c is the specific heat capacity of hot water, J / (kg·K); T s1,i The water supply temperature is ℃; T r1,i The return water temperature is ℃; T s2,i The water supply temperature for the secondary network is ℃; T r2,i The return water temperature of the secondary network is ℃; m 1.i For network traffic, kg / s; m 2.i ρ is the secondary network flow rate, kg / s; A is the heat exchange area, obtained from the heat exchanger sample, and is a constant, m. 2a and b are constants obtained by fitting historical data; the subscript i indicates time i.

[0023] Step S602: Based on the actual historical operating data at each time point, calculate the heat transfer coefficient k at each time point according to the heat exchanger model described above. i and heat exchange efficiency η i And based on its relationship with the network traffic m 1,i The parameters a and b are obtained by fitting the relationship.

[0024] Step S603: Set the network flow m at each time point obtained in step S5. 1,set,i The actual water supply temperature T s1,i Actual secondary network traffic m 2,i As input parameters, the particle swarm optimization algorithm is used to solve the above-mentioned heat exchanger physical model to calculate the corrected operating parameters at each time point, including the corrected primary return water temperature T. r1,corret,i Correct the secondary water supply temperature T s2,correct,i and correct the return water temperature T of the secondary network r2,correct,i The specific steps are as follows:

[0025] Step S6031: Correct the return water temperature T of the primary network. r1,corret,i and correct the return water temperature T of the secondary network r2,correct,i The particle swarm size and the number of iterations are set as variables.

[0026] Step S6032: Initialize the initial position and initial velocity of the particle swarm;

[0027] Step S6033: Calculate the objective function f for each particle, and find the optimal values ​​for both the individual and the group. The objective function is defined as follows:

[0028] Q 2,i,η =η i m 1,set,i (T s1,i -T r1,corret,i (9)

[0029]

[0030]

[0031] f = |Q 2,i,η -Q 2,i,ka |+|T sum2,set,i -T s2,correct,i -T r2,correct,i |×1000 (12)

[0032] In the formula: Q 2,i,η The heat exchange capacity of the secondary network is calculated using heat exchange efficiency, in W and T. r1,corret,iTo correct the return water temperature of the primary network, ℃; T s2,correct,i To correct the secondary water supply temperature, ℃; T r2,correct,i To correct the return water temperature of the secondary network, ℃; Q 2,i,ka W represents the heat exchange capacity of the secondary network, calculated using heat exchange efficiency and average temperature difference.

[0033] Step S6034: Update the velocity and position of each particle according to the following formula:

[0034]

[0035] In the formula: ω is the inertia factor; C1 and C2 are acceleration constants, generally C1 = C2 ∈ [0,4]; random(0,1) represents a random number on [0,1]. P id For the individual optimal solution, P gd This is the globally optimal solution.

[0036] Step S6035: The termination condition is whether the optimal objective function f is sufficiently close to zero. Determine if the termination condition is met. If so, the calculation ends, and the global optimal solution T is output. r1,corret,i and T r2,correct,i If not, repeat steps S6033-S6034 until the termination condition is met.

[0037] Step S604: Calculate the corrected heating network load Q at each time point. 1,set,i =cm 1,set,i (T s1,i -T r1,corret,i ).

[0038] Compared with the prior art, the advantages and positive effects of the present invention are:

[0039] (1) Use indoor temperature as feedback to correct the actual heating load. Considering that the actual data may have too much or too little heat supply, unnecessary energy waste can be avoided while ensuring indoor thermal comfort.

[0040] (2) The indoor temperature can be collected at a certain moment without installing sensors to continuously monitor the indoor temperature, which is economical and easier to implement.

[0041] (3) By combining the black box model and the white box model, a black box model and a white box model of the heat exchanger were established. The black box model obtains the appropriate flow rate of the network, and the white box model makes the corrected operating parameters conform to the actual physical meaning.

[0042] (4) Correcting the heating load based on indoor temperature can not only avoid overheating and undercooling, but also make physical corrections to the actual abnormal data based on physical characteristics, providing a solid data foundation for the next load forecast. Attached Figure Description

[0043] Figure 1 This is a flowchart of a heating load correction method based on indoor temperature.

[0044] Figure 2 The heating load correction results are shown in the example. Detailed Implementation

[0045] The present invention will now be described in more detail with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. It should be understood that those skilled in the art can modify the invention described herein while still achieving the beneficial effects of the invention. Therefore, the following description should be understood as being of broad knowledge to those skilled in the art and is not intended to limit the invention.

[0046] For clarity, not all features of the actual embodiments are described. In the following description, well-known functions and structures are not detailed in detail, as they would confuse the invention with unnecessary detail. It should be understood that in the development of any actual embodiment, numerous implementation details must be made to achieve the developer's specific objectives.

[0047] To make the objectives and features of this invention more apparent and understandable, the specific embodiments of this invention will be further described below with reference to the accompanying drawings. This application first provides a method for correcting heating load based on indoor temperature, including the following steps: Figure 1 As shown:

[0048] Step S1: Obtain the historical operating data of the heating station at various times and its corresponding outdoor meteorological parameters. The historical operating data includes the primary water supply temperature T. s1,i (°C), return water temperature of the first network T r1,i (°C), network traffic m 1,i (kg / s), secondary water supply temperature T s2,i (°C), Secondary network return water temperature T r2,i (°C), Secondary network traffic m 2,i (kg / s), outdoor meteorological parameters include outdoor dry-bulb temperature T out,i (°C). The subscript i indicates time i.

[0049] The example is the Agricultural Bank heating station in the Lanzhou New Area heating network. Table 1 lists its historical operating data and outdoor dry-bulb temperature at some points in time.

[0050] Table 1 Historical Operation Data of Agricultural Bank of China Heating Stations

[0051] Water supply temperature One-network return water temperature One-network traffic Secondary water supply temperature Secondary network return water temperature Secondary network traffic Collection time Outdoor dry bulb temperature 79.07805379 40.64067 6.272109 39.97155 34.30085 39.32478 2021 / 12 / 1 0:00 -8 80.04557114 40.59787 6.157251 39.96853 34.3316 39.2523 2021 / 12 / 1 1:00 -9 79.94249115 40.80946 6.342094 40.02279 34.28277 39.28267 2021 / 12 / 1 2:00 -9 79.99131775 40.6901 6.158383 39.94683 34.29904 39.25896 2021 / 12 / 1 3:00 -9 79.26323853 40.74978 6.319763 39.89258 34.18511 39.28552 2021 / 12 / 1 4:00 -10 79.75019243 40.99031 6.469952 40.05594 34.27071 39.30456 2021 / 12 / 1 5:00 -10 79.36197968 40.85287 6.349831 39.91971 34.17969 39.31164 2021 / 12 / 1 6:00 -10 79.42165756 41.24891 6.614411 40.12044 34.18511 39.27637 2021 / 12 / 1 7:00 -11 80.04014511 41.09701 6.456801 40.05534 34.16233 39.28048 2021 / 12 / 1 8:00 -11 80.25716095 41.23264 6.455963 40.08247 34.13845 39.31301 2021 / 12 / 1 9:00 -11 79.34027863 41.53646 6.926339 40.18555 34.09885 39.28282 2021 / 12 / 1 10:00 -7 79.17589111 40.98307 6.463658 39.8763 34.10916 39.28273 2021 / 12 / 1 11:00 -4 78.02598858 41.31266 6.721165 39.9821 34.1668 39.3206 2021 / 12 / 1 12:00 0 78.90828609 41.08073 6.391182 39.90072 34.22852 39.23908 2021 / 12 / 1 13:00 0 80.20833588 41.50468 5.475463 40.19717 34.29052 39.30649 2021 / 12 / 1 14:00 0 79.61154556 38.56156 4.998578 37.88339 33.29174 39.23559 2021 / 12 / 1 15:00 1 79.2550245 38.7269 4.912699 37.95961 33.37053 39.25307 2021 / 12 / 1 16:00 1 80.33854218 38.79666 4.903724 37.96116 33.25738 39.24071 2021 / 12 / 1 17:00 1 79.33123652 39.02271 5.204715 38.01903 33.19228 39.23147 2021 / 12 / 1 18:00 0 79.21617317 40.53819 6.145893 39.16965 33.63986 39.25484 2021 / 12 / 1 19:00 -1 79.5120786 40.21267 5.815999 38.96846 33.63534 39.32661 2021 / 12 / 1 20:00 -2 79.54824702 40.24884 5.784472 38.99559 33.67151 39.22445 2021 / 12 / 1 21:00 -2 79.31315231 40.26693 5.850905 38.96846 33.63534 39.22172 2021 / 12 / 1 22:00 -3 79.58441925 40.47038 5.867718 39.05436 33.66699 29.45439 2021 / 12 / 1 23:00 -5

[0052] Step S2: Collect the building's indoor temperature at a certain moment, denoted as T. in,0 (°C), and based on the current outdoor dry-bulb temperature T. out,0 Secondary water supply temperature T s2,0 Secondary network return water temperature T r2,0 According to the formula The heat exchange characteristic parameter BC0 of the building was calculated.

[0053] In this embodiment, the indoor temperature T at 12:00 on December 10, 2021 was collected. in,0 =24.55, current time T out,0 =1,T s2,0 =38.03711, T r2,0 =33.10547, therefore, BC0 = 0.468 is calculated.

[0054] Step S3: Based on the secondary network water supply temperature T at each time point s2,i Secondary network return water temperature T r2,i and outdoor dry-bulb temperature T out,i And the building heat transfer characteristic parameter BC0 obtained in step S2, to calculate the actual indoor temperature at all times.

[0055] The actual indoor temperatures calculated at various times of the day in the example are shown below. Figure 2 As shown by the curve.

[0056] Step S4: Set the indoor temperature at each time point to T according to actual needs. in,set,i According to formula T sum2,set,i =2BC0(T in,set,i -T out,i )+T in,set,i The sum of the set supply and return water temperatures (T) that should be achieved at each moment to meet the set indoor temperature is calculated. sum2,set,i .

[0057] In this embodiment, the indoor temperature T is set. in,set,i =20, BC0=0.468, according to formula T sum2,set,i =2BC0(T in,set,i -T out,i )+T in,set,i (3) The sum of the set supply and return water temperatures of the two networks at each time point, T, can be calculated. sum2,set,i For example, the T calculated at 12:00 on December 10, 2021. sum2,set,0 =57.784.

[0058] Step S5: Establish a support vector regression (SVR) black box model of the heat exchanger to obtain the set indoor temperature T. in,set,i The required operating parameters for the heating network are set as follows: Network flow rate (m³) 1,set,i .

[0059] Step S501: Based on the actual operating data of the heating station at each time, a support vector regression (SVR) model for the heat exchanger is trained. The input parameters are the primary network supply water temperature, secondary network flow rate, secondary network supply water temperature and secondary network return water temperature at each time. The output parameter is the primary network flow rate at the corresponding time.

[0060] Step S502: Maintain the set temperature difference of the secondary network at each moment to be the same as the actual operating data, i.e., T s2,set,i,5 -T r2,set,i,5 =T s2,,i -T r2,i The sum of the supply and return water temperatures of the secondary network at each time point obtained according to step S4 is T. sum2,set,i It can solve for the set water supply temperature T of the secondary network at each moment. s2,set,i,5 The set return water temperature T of the secondary network r2,set,i,5 .

[0061] Step S503: Calculate the actual water supply temperature T at each time point. s1,i Actual secondary network traffic m 2,i Set the secondary water supply temperature T s2,set,i,5 Set the secondary network return water temperature T r2,set,i,5 Using the support vector regression (SVR) model of the heat exchanger as input data, the set primary flow rate m at each time step is obtained. 1,set,i .

[0062] In this embodiment, a heat exchanger SVR model was established based on actual historical operating data according to the above steps, and T was made... s2,set,i,5 -T r2,set,i,5 =T s2,,i -T r2,i The set water supply temperature T of the secondary network at various times was calculated. s2,set,i,5 The set return water temperature T of the secondary network r2,set,i,5 Ultimately, the actual water supply temperature T of the entire network will be... s1,i Actual secondary network traffic m 2,i Set the secondary water supply temperature T s2,set,i,5 Set the secondary network return water temperature T r2,set,i,5 Using the support vector regression (SVR) model of the heat exchanger as input data, the set primary flow rate m at each time step is obtained. 1,set,i Taking 12:00 on December 10, 2021 as an example, the calculated network flow m is... 1,set,0 = 4.47206, while the original network traffic m 1,0=5.825211. Since the current indoor temperature exceeds the set indoor temperature, it indicates that there is overheating. Therefore, the calculated set flow rate of the primary network is less than the actual flow rate of the primary network, which is consistent with physical laws.

[0063] Step S6: Based on the physical model of the heat exchanger white box, and according to the set primary flow rate m... 1,set,i Solve for the corrected network load Q 1,set,i .

[0064] Step S601: Establish the physical model of the heat exchanger according to the following formula.

[0065] Q 1,i =cm 1.i (T s1,i -T r1,i (4)

[0066] Q 2.i =cm 2.i (T s2,i -T r2,i (5)

[0067] η i ×Q 1,i =Q 2.i (6)

[0068]

[0069] k i =am 1.i +b (8)

[0070] In the formula, Q 1,i For heat exchange in a single network, W; Q 2.i For heat exchange in the secondary network, W; η i For heat exchange efficiency, k i The heat transfer coefficient is W / (m²). 2 ·K); c is the specific heat capacity of hot water, J / (kg·K); T s1,i The water supply temperature is ℃; T r1,i The return water temperature is ℃; T s2,i The water supply temperature for the secondary network is ℃; T r2,i The return water temperature of the secondary network is ℃; m 1.i For network traffic, kg / s; m 2.i ρ is the secondary network flow rate, kg / s; A is the heat exchange area, obtained from the heat exchanger sample, and is a constant, m. 2 a and b are constants obtained by fitting historical data; the subscript i indicates time i.

[0071] Step S602: Based on the actual historical operating data at each time point, calculate the heat transfer coefficient k at each time point according to the heat exchanger model described above.i and heat exchange efficiency η i And based on its relationship with the network traffic m 1,i The parameters a and b are obtained by fitting the relationship.

[0072] Step S603: Set the network flow m at each time point obtained in step S5. 1,set,i The actual water supply temperature T s1,i Actual secondary network traffic m 2,i Using particle swarm optimization as input parameters, the physical model of the heat exchanger is solved to calculate the corrected primary return water temperature T at each time point. r1,corret,i Correct the secondary water supply temperature T s2,correct,i and correct the return water temperature T of the secondary network r2,correct,i The specific steps are as follows:

[0073] Step S6031: Correct the return water temperature T of the primary network. r1,corret,i and correct the return water temperature T of the secondary network r2,correct,i The particle swarm size and the number of iterations are set as variables.

[0074] Step S6032: Initialize the initial position and initial velocity of the particle swarm;

[0075] Step S6033: Calculate the objective function f for each particle, and find the optimal values ​​for both the individual and the group. The objective function is defined as follows:

[0076] Q 2,i,η =η i m 1,set,i (T s1,i -T r1,corret,i (9)

[0077]

[0078]

[0079] f = |Q 2,i,η -Q 2,i,ka |+|T sum2,set,i -T s2,correct,i -T r2,correct,i |×1000 (12)

[0080] In the formula: Q 2,i,η The heat exchange capacity of the secondary network is calculated using heat exchange efficiency, in W and T. r1,corret,i To correct the return water temperature of the primary network, ℃; T s2,correct,i To correct the secondary water supply temperature, ℃; T r2,correct,i To correct the return water temperature of the secondary network, ℃; Q 2,i,ka W represents the heat exchange capacity of the secondary network, calculated using heat exchange efficiency and average temperature difference.

[0081] Step S6034: Update the velocity and position of each particle according to the following formula:

[0082]

[0083] In the formula: ω is the inertia factor; C1 and C2 are acceleration constants, generally C1 = C2 ∈ [0,4]; random(0,1) represents a random number on [0,1]. P id For the individual optimal solution, P gd This is the globally optimal solution.

[0084] Step S6035: The termination condition is whether the optimal objective function f is sufficiently close to zero. Determine if the termination condition is met. If so, the calculation ends, and the global optimal solution T is output. r1,corret,i and T r2,correct,i If not, repeat steps S6033-S6034 until the termination condition is met.

[0085] Step S604: Calculate the corrected heating network load Q at each time point. 1,set,i =cm 1,set,i (T s1,i -T r1,corret,i ).

[0086] In this embodiment, taking 12:00 on December 10, 2021 as an example, the calculated network traffic m is... 1,set,0 =4.47206, the heat transfer coefficient k obtained by fitting according to the above steps i With one network traffic m 1,i The coefficients in the relation are a = 103.002275 and b = 212.846718. Based on the white-box physical model of the heat exchanger, T is obtained using the particle swarm optimization algorithm. r1,corret,i =33.34915, T r2,correct,i =26.5807, the corrected heat load Q is calculated. 1,set,i =241.97KW. Original heat load Q of the primary power grid before correction. 1,i =258.15KW. From the results, it can be seen that under the condition of overheating, the corrected heat load is lower than the original heat load of the primary network. Therefore, it is more reasonable to make load prediction based on the corrected heat load data. Figure 2 The load correction results for a 24-hour period on December 9, 2021, are shown in the example.

Claims

1. A method for correcting a heating load based on an indoor temperature, characterized by, The method comprises the following steps: Step S1: obtaining historical operation data of the heat station at each time and corresponding outdoor meteorological parameters; Step S2: collect the indoor temperature of the building at a certain time and calculate the building heat exchange characteristic parameters according to the outdoor dry bulb temperature and the supply and return water temperature of the second network at the current time ; Step S3: according to the water supply and return water temperature of each time and the outdoor dry bulb temperature, and the building heat exchange characteristic parameters calculated in step S2 , the actual indoor temperature at all times is calculated; Step S4: Set the indoor temperature at each time according to actual needs , and calculate the sum of the set supply and return water temperatures at each time to meet the set indoor temperature . Step S5: Establishing a heat exchanger support vector regression (SVR) black box model to obtain the set indoor temperature The operating parameter setting of the lower heat network required is a network flow ; Step S6: According to the heat exchanger white box physical model, in the case of setting a network flow Solve the modified one network and two network operation parameters, and calculate the modified one network load .

2. The heating load correction method based on indoor temperature according to claim 1, characterized in that: In step S1, the historical operation data of the heat station at each time and the corresponding outdoor meteorological parameters are acquired, the historical operation data including one-network water supply temperature (℃), one-network return water temperature (℃), one-network flow ( ), two-network water supply temperature (℃), two-network return water temperature (℃), two-network flow ( ), and the outdoor meteorological parameters including outdoor dry-bulb temperature (℃), and subscript represents time.

3. The heating load correction method based on indoor temperature according to claim 1, characterized in that: In step S2, the building heat exchange characteristic parameter is calculated according to the following formula : （1） In the formula: is the indoor temperature of the building at the time of collection, ℃; is the outdoor dry-bulb temperature at the time, ℃; is the temperature of the secondary network water supply at the time, ℃; is the temperature of the secondary network return water at the time, ℃.

4. The heating load correction method based on indoor temperature according to claim 1, characterized in that: In step S3, the actual indoor temperature at all times is calculated according to the following formula: （2） In the formula: is the actual indoor temperature at the moment, ℃; is the outdoor dry-bulb temperature at the moment, ℃; is the actual indoor temperature at the moment, ℃; is the outdoor dry-bulb temperature at the moment, ℃; is the two-network supply water temperature at the moment, ℃; is the two-network supply water temperature at the moment, ℃; is the two-network return water temperature at the moment, ℃; is the two-network return water temperature at the moment, ℃; is the building heat exchange characteristic parameter.

5. The indoor temperature-based heating load correction method of claim 1, wherein: In step S4, the sum of the set supply and return water temperatures of the two networks that should be reached at each time to satisfy the set indoor temperature is calculated according to the following equation : （3） In the formula: is the set indoor temperature at the time, °C is the outdoor dry-bulb temperature at the time, °C.

6. The indoor temperature-based heating load correction method of claim 1, wherein: In step S5, based on the actual operation data of the heat station at each time, a heat exchanger support vector regression (SVR) model is trained, the input parameters are the one-network water supply temperature, the two-network flow, the two-network water supply temperature and the two-network return water temperature at each time, and the output parameter is the one-network flow at the corresponding time; the two-network set temperature difference at each time is kept the same as the actual operation data, that is , and the sum of the two-network water supply and return water temperatures at each time obtained in step S4 , the two-network set water supply temperature at each time is obtained and the two-network set return water temperature ; the actual one-network water supply temperature at each time , the actual two-network flow , the set two-network water supply temperature , and the set two-network return water temperature are taken as the input data of the heat exchanger support vector regression (SVR) model, and the set one-network flow at each time is obtained .

7. The method of claim 1, wherein the indoor temperature-based heating load correction method is characterized by: In step S6, a white-box physical model of the heat exchanger is established according to the following formula: （4） （5） （6） （7） （8） wherein, is the heat exchange amount of the first network, ; is the heat exchange amount of the second network, is the heat exchange efficiency, is the heat exchange coefficient, ; is the specific heat capacity of hot water, ; is the water supply temperature of the first network, ℃; is the return water temperature of the first network, ℃; is the water supply temperature of the second network, ℃; is the return water temperature of the second network, ℃; is the flow of the first network, ; is the flow of the second network, ; is the heat exchange area, obtained according to a heat exchanger sample, is a constant, ; , is a constant, obtained by fitting historical data; the subscript represents the moment.

8. The indoor temperature-based heating load correction method of claim 7, wherein: In step S6, the set one-net flow rate at each time is calculated , the actual one-net water supply temperature , the actual two-net flow rate As input parameters, the particle swarm algorithm is used to solve the white-box physical model of the heat exchanger, and the corrected one-net return water temperature at each time is calculated .

9. The method for correcting a heating load based on an indoor temperature according to claim 8, characterized by: In step S6, in the solving process using the particle swarm algorithm, the corrected primary network return water temperature and the corrected secondary network return water temperature are taken as variables, and the objective function of each particle is defined as: （9） （10） （11） + （12） In the formula: Q2 is the heat exchange amount of the second network calculated by the heat exchange efficiency, T1 is the return water temperature of the first network, ℃; T2 is the supply water temperature of the second network, ℃; T3 is the return water temperature of the second network, ℃; Q2 is the heat exchange amount of the second network calculated by the heat exchange efficiency and the average temperature difference, .​

Images