Automatic operation and regulation method of heating system based on characteristic parameter identification

Abstract

The present invention relates to the field of heating, and specifically to an automatic operation and regulation method of a heating system based on characteristic parameter identification. The present invention lists the process description equations of the primary side heating supply of a heat exchange station, the secondary side heating supply of a heat exchange station, the heat exchange of a heat exchanger, the heat dissipation of a user radiator, and the heat consumption of a user enclosure structure, finds out the characteristic constants that characterize the operating characteristics of a heating system, takes the historical data of seven parameters, namely, the primary side water supply temperature of a heat exchange station, the primary side return water temperature of a heat exchange station, the primary side flow rate of a heat exchange station, the secondary side water supply temperature of a heat exchange station, the secondary side return water temperature of a heat exchange station, the user's indoor temperature, and the outdoor meteorological temperature as input, identifies the characteristic constants of a heating system through a maximum likelihood estimation method, obtains the specific equations of each heat transfer process of the entire heating system, and completes the heat load prediction, the determination of the heat source water supply temperature, and the primary network flow distribution on this basis, so that the prediction and control model is closer to the actual operating law of the heating system.

Description

Technical Field

[0001] The present invention relates to the field of heating, and in particular to an automatic operation adjustment method for a heating system based on characteristic parameter identification. Background Art

[0002] For a long time, under the extensive management model, the operation and adjustment of the heating system was determined by the operation and maintenance personnel based on past experience. The heat supply of the heat source often deviated from the actual heat load demand, resulting in energy waste and increased emissions of atmospheric pollutants. The heat source water supply temperature was determined based on the theoretical calculation of the heating temperature adjustment curve, which was often inconsistent with the actual operation needs. The opening of the primary-side regulating valve of the heat exchanger station was manually determined by the operation and maintenance personnel based on the primary-side return water temperature of each heat exchanger station. The workload of network adjustment was large, and hydraulic imbalance often occurred.

[0003] With the development of technologies such as the Internet of Things and information systems, data analysis, and artificial intelligence, my country has begun to pay attention to and study smart heating in order to achieve refined, automated operation and regulation of the heating system and achieve the goal of energy conservation and consumption reduction, focusing on solving problems such as heat load prediction of the heating system, determination of heat source water supply temperature, and primary network flow distribution.

[0004] At present, the commonly used methods for predicting the heat load of the heating system, determining the heat source water supply temperature, and allocating the primary network flow are as follows:

[0005] 1. Common methods for heat load prediction

[0006] a. Traditional heat load prediction methods

[0007] The similar day method is a typical traditional heat load forecasting method, which makes predictions based on historical heat load data with similar meteorological and working day attributes.

[0008] b. Prediction method based on heat load series

[0009] Time series method and grey theory method are the main prediction methods based on heat load series.

[0010] The time series method is to first decompose seasonal changes from the time series, find out the rules of seasonal changes, and then make predictions based on the long-term trends.

[0011] Grey theory abandons the link of system structure analysis, directly seeks the overall law of the system by accumulating the original data, and constructs an exponential model. Grey theory heat load prediction is based on the differential equation G(1,1) model of order 1 and 1 variable, which has the characteristics of small model data volume, simplicity and convenience. On the basis of the grey model, the residual sequence obtained from the grey model is divided into multiple periodic waves, and each periodic wave is extended as needed, and finally each periodic wave is linearly superimposed to obtain an improved grey model, which can reflect the seasonality and periodicity of heat load changes.

[0012] c. Prediction methods based on artificial intelligence

[0013] Neural networks and support vector machines are relatively mature artificial intelligence heat load prediction methods.

[0014] The neural network method simulates the structure, characteristics and thinking mode of artificial neural networks, uses learning and training of empirical samples to continuously adjust the connection weights, and finally establishes a nonlinear mapping between multiple input variables and multiple output variables. Its biggest feature is that it does not require the establishment of specific specifications and mathematical models.

[0015] The support vector machine method transforms the input space X into a high-dimensional space through a nonlinear transformation defined by the inner product kernel function K(X), and then finds the optimal classification surface in this space. Its output Y is a linear combination of intermediate nodes, and each intermediate node corresponds to the inner product of an input sample and a support vector. This model does not have the problem of local minimum value and has obvious advantages in solving complex problems such as fuzzy, random, uncertain, limited sample number and nonlinearity.

[0016] 2. Common methods for determining heat source water supply temperature

[0017] a. Determined by theoretical calculation of heating regulation curve

[0018] According to the basic heating equation, the curve of the heat source water supply temperature changing with the outdoor temperature under different adjustment methods can be calculated and determined. The operator can directly adjust the heat source water supply temperature according to the outdoor temperature and the given temperature adjustment curve. This method is simple and convenient, but due to the deviation between the actual operating conditions and the design conditions, the actual heating effect is inconsistent with the design value.

[0019] b. Determine the water supply temperature of the heat source using the similar day method

[0020] This method is determined based on historical data of heat source water supply temperature with similar meteorological and working day attributes, which solves the deviation between theoretical calculation and actual operation, but the operation and adjustment mode of the heating system should remain consistent.

[0021] 3. Common methods for primary network traffic distribution

[0022] In the traditional manual adjustment method, the operation and maintenance personnel reduce the opening of the regulating valve of the heat exchange station with higher return water temperature on the primary side, and increase the opening of the regulating valve of the heat exchange station with lower return water temperature on the primary side, so that the supply and return water temperature difference of each heat exchange station is basically equal, and the primary network flow is "distributed on demand". The actual operation of this method is complicated, and the opening of the regulating valve needs to be adjusted repeatedly, which is labor-intensive.

[0023] After the unmanned transformation of the heat exchange station, the opening of the regulating valve can be automatically controlled by the control system according to the return water temperature, realizing the transition from manual to automatic control, but the regulation principle has not changed. Since the return water temperature change has a lag, the adjustment time is relatively long and the regulating valve still needs to be adjusted repeatedly.

[0024] The common methods for predicting heat load of heating system, determining heat source water supply temperature and allocating primary network flow still have the following problems:

[0025] 1. The currently used heat load prediction methods have all abandoned the variable analysis link of the heating system, and directly found the nonlinear relationship between the heat load and various influencing factors through big data and artificial intelligence algorithms, using mature mathematical prediction models. However, there are more or less differences between the relationship between all mathematical prediction models and the variables of the actual heating system, which not only makes the prediction results have certain errors, but also cannot truly reflect the heat transfer process between the heat source, heat exchange station and heat users of the heating system.

[0026] 2. There is a deviation between the heat source water supply temperature calculated by the basic heating formula theory and the actual operation requirements. When the heat source water supply temperature is determined by the similar day method, it is required that the heating system operates in the same way under similar meteorological and working day conditions.

[0027] 3. The primary network flow distribution and regulation system is based on the return water temperature value on the primary side of the heat exchange station. The regulating valve needs to be adjusted repeatedly. Since the return water temperature change has a lag, the adjustment time is relatively long. Summary of the invention

[0028] The purpose of the present invention is to solve the technical problems that the prediction model in the current heat load prediction process cannot truly reflect the heat transfer process between the heat source, heat exchange station and heat users of the heating system, there is a deviation between the theoretical calculation value of the heat source water supply temperature and the actual demand, and the adjustment time is long in the process of primary network flow distribution adjustment based solely on the return water temperature. A method for automatic operation and adjustment of a heating system based on characteristic parameter identification is provided, which is used for thermal load prediction, heat source water supply temperature determination and primary network flow distribution of smart heating, so as to realize "heating on demand" and "distribution on demand" of the heating system and achieve the purpose of energy saving and environmental protection.

[0029] The technical solution adopted by the present invention to achieve the above-mentioned purpose is: a method for automatically adjusting the operation of a heating system based on characteristic parameter identification, comprising the following steps:

[0030] S1: List the process description equations of the primary side heat supply of the heat exchange station, the secondary side heat supply of the heat exchange station, the heat exchange of the heat exchanger, the heat dissipation of the user radiator, and the heat consumption of the user's enclosure structure, find out the characteristic constants that characterize the operating characteristics of the heating system, and convert the primary side water supply temperature T g_1 , Heat exchange station primary side return water temperature T h_1, primary side flow rate F of heat exchange station g_1 , Heat exchange station secondary side water supply temperature T g_2 , Heat exchange station secondary side return water temperature T h_2 , User indoor temperature T n and outdoor weather temperature T w The historical data of seven parameters are used as input, and the characteristic constants of the heating system are identified through the maximum likelihood estimation method, and the specific equations of each heat transfer process of the entire heating system are obtained;

[0031] S2: Based on the identification of characteristic parameters of the heating system, a heat load prediction model, a primary network flow distribution model, a heat source water supply temperature recommended value calculation model, a user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and a heat exchange station heat consumption evaluation model are constructed. The heat load of the entire heating system is predicted by the load prediction model, the primary network distribution flow of each heat exchange station is determined by the primary network flow distribution model, the heat source water supply temperature is determined by the heat source water supply temperature recommended value calculation model, the user indoor temperature average value within the heating area of ​​the heat exchange station is estimated by the user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and the heat consumption evaluation model of the heat exchange station is used to evaluate the heat consumption of the heat exchange station.

[0032] Furthermore, the step S1 includes the following steps:

[0033] S11: Constructing the maximum likelihood estimation regression theory model:

[0034] 1) The maximum likelihood estimation model of the parameters of the two-variable linear regression model can be expressed as follows:

[0035] y=αx+β+ε (1)

[0036] In the formula: y is the response variable; x is the dependent variable; α is the slope of the line; β is the intercept of the line, and ε is the error;

[0037] The maximum likelihood estimation results of the model parameters are as follows:

[0038]

[0039] Where: is the sample mean of the response variable; is the sample mean of the dependent variable; is the maximum likelihood estimate of the slope of the straight line; is the maximum likelihood estimate of the line intercept;

[0040]

[0041] Where: x i is the dependent variable in the i-th group of samples; y i is the response variable in the i-th group of samples;

[0042]

[0043] Where: is the variance estimate; n is the sample size; ε i is the error in the i-th group of samples;

[0044] 2) The maximum likelihood estimation model of the parameters of the multivariate linear regression model can be expressed as follows:

[0045] y=β0+β1x1+β2x2+...+β k x k +ε (5)

[0046] Where: y is the response variable; β0 is the spatial intercept; x1, x2, x k is the dependent variable related to the response variable; β1, β2, β k are the dependent variables x1, x2, x k The corresponding weight factor; ε is the error;

[0047] For a model with n sets of measurements, the matrix expression is:

[0048] Y=Xβ+E (6)

[0049] In the formula: Y is the response variable sample matrix; X is the dependent variable sample matrix; β is the dependent variable weight factor matrix; E is the sample error matrix;

[0050]

[0051]

[0052] The maximum likelihood estimation results of the model parameters are as follows:

[0053]

[0054] Where: is the maximum likelihood estimate of the dependent variable weight factor matrix;

[0055]

[0056] Where: is the variance estimate; n is the sample size;

[0057] S12: List the basic formulas for heat exchange stations and user heating:

[0058] The primary network heat supply can be expressed as follows:

[0059] Q g_1 =C1·(T g_1 -T h_1 )·Fg_1 (13)

[0060] Where: Q g_1 is the heat supply of the primary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_1 , T h_1 are the primary side supply and return water temperatures of the heat exchange station, ℃; F g_1 is the water supply flow rate on the primary side of the heat exchange station, t / h;

[0061] The secondary network heat supply can be expressed as follows:

[0062] Q g_2 =C1·(T g_2 -T h_2 )·F g_2 (14)

[0063] Where: Q g_2 is the heat supply of the secondary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_2 , T h_2 are the secondary side supply and return water temperatures of the heat exchange station, ℃; F g_2 is the water supply flow rate on the secondary side of the heat exchange station, t / h;

[0064] When the water on the primary and secondary sides of the heat exchanger flows in reverse, the heat exchange between the primary and secondary sides of the heat exchanger can be expressed as follows:

[0065]

[0066] Where: Q 1_2 is the heat exchange capacity of the primary and secondary sides of the heat exchange station, kw; C 1-2 is a static constant related to the heat transfer coefficient and heat transfer area of ​​the heat exchanger;

[0067] When the heat transfer coefficient of the radiator is a constant, the heat dissipation of the radiator can be expressed as follows:

[0068]

[0069] Where: Q s_s is the heat dissipation of the radiator, kw; C2 is a static constant related to the heat transfer coefficient and radiator area of ​​the radiator; T n is the user's indoor temperature, ℃;

[0070] The heat consumption of a building is mainly related to the heat transfer coefficient of the building envelope, the area of ​​the building envelope, the indoor temperature and the outdoor temperature. When the heat transfer coefficient of the building envelope is a constant, the heat consumption of the building can be expressed as follows:

[0071] Q j_h =C 3_1 ·[T n -(C3_2 ·T w +C 3_3 )] (17)

[0072] Where: Q j_h is the heat consumption of the building envelope, kw; C 3_1 is a static constant related to the heat transfer coefficient and area of ​​the enclosure structure; T w is the outdoor temperature from meteorological data, ℃; 3_2 , C 3_3 is the correction constant for the outdoor temperature in the meteorological data;

[0073] S13: Count the primary side water supply temperature T of each heat exchange station g_1 , primary side return water temperature T h_1 , primary side flow F g_1 , secondary side water supply temperature T g_2 , secondary side return water temperature T h_2 , User indoor temperature T n Historical operation data and the outdoor temperature forecast value T at the corresponding time w ;

[0074] S14: Identification of characteristic parameters of heating system:

[0075] 1) C1 is related to the specific heat capacity of water. According to the principle of dimensional harmony, C1 is 1.1625;

[0076] 2) Identify C by using the two-variable linear regression model in step S11 1_2 :

[0077] C 1_2 Identify the maximum likelihood estimation model:

[0078]

[0079] Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement;

[0080] 3) Identify C2 based on the two-variable linear regression model in step S11:

[0081] C2 identification maximum likelihood estimation model:

[0082]

[0083] Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement;

[0084] 4) Identify 1 / C by using the multivariate linear regression model in step S113_1 +1 / C2, C 3_2 , C 3_3 :

[0085] 1 / C 3_1 +1 / C2, C 3_2 , C 3_3 Identify the maximum likelihood estimation model:

[0086]

[0087] Where: ε is the random error caused by the system and measurement;

[0088] C 3_1 Based on the C2 identification result in step 3) and 1 / C 3_1 +1 / C2 identification results show that:

[0089]

[0090] Furthermore, in step S2, a heat load prediction model is constructed based on the identification of characteristic parameters of the heating system, including the following steps:

[0091] (1) Given the external environmental variables T of each heat exchange station w (n);

[0092] (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n);

[0093] (3) Predict the heat load Q0(n) of each heat exchange station:

[0094] The heat load prediction formula for each heat exchange station is as follows:

[0095] Q0(n)=C 3_1 (n)·{T n (n)-[C 3_2 (n)·T w (n)+C 3_3 (n)]} (22)

[0096] Where: Q0(n) is the predicted value of heat load of each heat exchange station, kw; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the meteorological data of each heat exchange station;

[0097] (4) Predict the heat load Q0 of the entire heating system:

[0098] The heat load of the entire heating system is the sum of the heat loads of each heat exchange station:

[0099] Q0=∑Q0(n) (23).

[0100] Furthermore, in step S2, a primary network flow distribution model is constructed based on the identification of characteristic parameters of the heating system, including the following steps:

[0101] (1) Given the external environmental variables T of each heat exchange station at the same time w (n);

[0102] (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n);

[0103] (3) Calculate the total flow rate F of the primary network of the heat source g_0 ;

[0104] (4) Determine the primary network distribution flow F of each heat exchange station g_1 (n):

[0105] The calculation formula for the primary network flow of each heat exchange station is as follows:

[0106]

[0107] Where: C 3_1 is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and area of ​​the enclosure structure; C 3_2 , C 3_3 is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the relevant meteorological data; n and m represent heat exchange station n and heat exchange station m respectively.

[0108] Furthermore, in step S2, a heat source water supply temperature recommendation value calculation model is constructed based on the identification of characteristic parameters of the heating system, including the following steps:

[0109] (1) Given the external environmental variables T of the heat exchange station w (n);

[0110] (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n);

[0111] (3) Set the expected value T of the indoor temperature in the heating area of ​​the heat exchange station n (n);

[0112] (4) Give the recommended value of heat source water supply temperature T g_0 :

[0113] The heat source water supply temperature is obtained by solving the following implicit function:

[0114]

[0115] Where: C1(n) is 1.1625; C2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and radiator area of ​​each heat exchange station; C 1_2 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and heat exchange area of ​​each heat exchange station; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system based on the outdoor temperature in the meteorological data of each heat exchange station.

[0116] Furthermore, in step S2, a model for estimating the average indoor temperature of users in the heating area of ​​the heat exchange station is constructed based on the identification of characteristic parameters of the heating system, including the following steps:

[0117] (1) Given the external environmental variables T of the heat exchange station w (n);

[0118] (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n);

[0119] (3) Given the heat source water supply temperature T g_0 ;

[0120] (4) Estimate the average room temperature T of users in the heating area of ​​the heat exchange station by solving the following implicit function formula (25) n (n):

[0121]

[0122]

[0123] Where: C1(n) is 1.1625; C2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and radiator area of ​​each heat exchange station; C 1_2 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and heat exchange area of ​​each heat exchange station; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system based on the outdoor temperature in the meteorological data of each heat exchange station.

[0124] Furthermore, in step S2, a heat exchange station heat consumption evaluation model is constructed based on the identification of characteristic parameters of the heating system, which is divided into the following two cases:

[0125] 1) Heat consumption evaluation of different heat exchange stations in the same heating season includes the following steps:

[0126] (1) Statistics on the heating area F(n) and the cumulative annual heating supply of each heat exchange station The cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station;

[0127] (2) Calculate the annual heat supply per unit heating area of ​​each heat exchange station

[0128] The calculation formula for the annual heat supply per unit heating area is as follows:

[0129]

[0130] (3) Compare the heat supply per unit area of ​​each heat exchange station and evaluate the heat consumption;

[0131] 2) The heat consumption evaluation of the same heat exchange station in different heating seasons includes the following steps:

[0132] (1) Calculate the heating area F(n) and the cumulative annual heating supply of each heat exchange station and heating degree days HDD18, the cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station;

[0133] (2) Calculate the heat consumption per square meter of heating area per heating degree-day of each heat exchange station

[0134] The calculation formula of heat consumption per square meter of heating area per unit heating degree day is as follows:

[0135]

[0136] (3) Compare the heat consumption per square meter of heating area per heating degree-day of each heat exchange station and conduct heat consumption evaluation.

[0137] Compared with the prior art, the present invention has the following beneficial effects:

[0138] (1) In the process of operation and management of the heating system, the heating enterprise of the present invention stores a large amount of historical operation data, which reflects the potential operation characteristics of the heating system, making it possible to use machine learning algorithms to identify the characteristic parameters of the heating system and to establish a heat load prediction and heating network automatic operation adjustment model based on this.

[0139] (2) The present invention predicts heat load and automatically adjusts operation based on the identification of characteristic parameters of the heating system. Guided by theoretical heat transfer formulas, the prediction and control models are closer to the actual operation rules of the heating system, reducing prediction errors and further reducing the amount of sample data, thus achieving "on-demand heating" of the heating system heat source and "on-demand distribution" of the primary network flow of each heat exchange station, thus achieving the goal of energy saving and environmental protection.

[0140] (3) The prediction model in the heat load prediction process of the present invention can truly reflect the problems of heat transfer process between the heat source, heat exchange station and heat users of the heating system, and can reduce the deviation between the theoretical calculated value of the heat source water supply temperature and the actual demand, thus solving the problem of long adjustment time in the process of adjusting the primary network flow distribution by relying solely on the return water temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0141] Figure 1 This is a structural diagram of the heating automatic operation and regulation system based on characteristic parameter identification of the present invention.

[0142] In the figure: 1. Heat source water supply temperature sensor, 2. Heat source return water temperature sensor, 3. Heat source return water flow meter, 4. Heat exchange station primary side water supply temperature sensor, 5. Heat exchange station primary side return water temperature sensor, 6. Heat exchange station primary side electric regulating valve, 7. Heat exchange station primary side flow meter, 8. Heat exchange station secondary side water supply temperature sensor, 9. Heat exchange station secondary side return water temperature sensor, 10. Heat exchange station secondary side water supply flow meter, 11. Heat exchanger, 12. Radiator, 13. Room temperature sensor. DETAILED DESCRIPTION

[0143] The present invention is described in detail below with reference to the accompanying drawings and embodiments, but the present invention is not limited to the specific embodiments.

[0144] like Figure 1 The heating automatic operation and regulation system based on characteristic parameter identification shown in the figure builds a heat exchange station between the heat source and each heating area, a primary network management network is built between the heat source and the heat exchange station, a secondary network management network is built between the heat exchange station and the heating users, and a radiator 12 is installed in the indoor room of the heating user.

[0145] A system management platform and a data storage server are provided in the control center. The system management platform and the data storage server are signal-connected. The input end of the system management platform is respectively signal-connected to the user side room temperature sensor 13, the heat source water supply temperature sensor 1, the heat source return water temperature sensor 2, the heat source return water flowmeter 3, the heat exchange station primary side water supply temperature sensor 4, the heat exchange station primary side return water temperature sensor 5, the heat exchange station primary side flowmeter 7, the heat exchange station secondary side water supply temperature sensor 8, the heat exchange station secondary side return water temperature sensor 9 and the heat exchange station secondary side water supply flowmeter 10. The output end of the system management platform is signal-connected to the input end of the PLC controller of the primary side electric regulating valve of the heat exchange station, and the output end of the PLC controller of the primary side electric regulating valve of the heat exchange station is signal-connected to the primary side electric regulating valve 6 of the heat exchange station.

[0146] The signal transmission method between the user-side room temperature sensor 13 and the system management platform is wireless transmission. The user-side room temperature sensor 13 adopts a C15W-NB crystal vibration type digital temperature sensor; the heat source supply and return water temperature sensor, the heat exchange station primary side supply and return water temperature sensor and the heat exchange station secondary side supply and return water temperature sensor all adopt PT100 temperature sensor; the heat source return water flowmeter 3, the heat exchange station primary side flowmeter 7 and the heat exchange station secondary side supply water flowmeter 10 all adopt electromagnetic flowmeters or ultrasonic flowmeters.

[0147] The heat source water supply temperature sensor 1 is installed on the water supply pipe at the heat source outlet, and the heat source return water temperature sensor 2 is installed on the return water pipe at the heat source inlet, which is used to measure the heat source supply and return water temperatures. The heat source return water flowmeter 3 is installed on the straight pipe section of the heat source return water pipe, and there is a straight pipe section with a length of at least 10 times the pipe diameter at the front end of the flowmeter, and a straight pipe section with a length of at least 5 times the pipe diameter at the rear end of the flowmeter, which is used to measure the heat source return water flow.

[0148] The heat exchange station is provided with a heat exchanger 11, the primary side water supply temperature sensor 4 of the heat exchange station is installed on the primary side water supply pipe of the heat exchange station, the primary side return water temperature sensor 5 of the heat exchange station is installed on the primary side return water pipe of the heat exchange station, and is used to measure the primary side supply and return water temperature of the heat exchange station, the primary side flow meter 7 of the heat exchange station is installed on the straight pipe section of the primary side return water pipe of the heat exchange station, and the front end of the flow meter has a straight pipe section of at least 10 times the pipe diameter length, and the rear end of the flow meter has a straight pipe section of at least 5 times the pipe diameter length, which is used to measure the primary side flow of the heat exchange station, the primary side flow meter 7 of the heat exchange station can also be installed on the primary side water supply pipe of the heat exchange station, and the primary side electric regulating valve of the heat exchange station 6 is installed on the primary side water supply pipeline of the heat exchange station, and the PLC controller executes the operation adjustment instructions of the system management platform to control the opening of the electric regulating valve; the secondary side water supply temperature sensor 8 of the heat exchange station is installed on the secondary side water supply pipeline of the heat exchange station, and the secondary side return water temperature sensor 9 of the heat exchange station is installed on the secondary side return water pipeline of the heat exchange station, which is used to measure the secondary side supply and return water temperature of the heat exchange station, and the secondary side water supply flowmeter 10 of the heat exchange station is installed on the straight pipe section of the secondary side water supply pipeline of the heat exchange station, and the front end of the flowmeter has a straight pipe section of at least 10 times the pipe diameter length, and the rear end of the flowmeter has a straight pipe section of at least 5 times the pipe diameter length, which is used to measure the secondary side flow of the heat exchange station.

[0149] The system management platform integrates the automatic operation and adjustment model algorithm of the heating system. Through characteristic parameter identification, the primary network flow of the heat exchange station is allocated on demand to form the adjustment instructions of the electric control valve on the primary side of the heat exchange station. The system management platform can display and query the operating parameters of the heating system, the predicted value of the heat load of the heating system, the recommended value of the heat source water supply temperature and the energy consumption of each heat exchange station. The data storage device stores historical operating data and meteorological data for calling the automatic operation and adjustment model algorithm of the heating system in the system management platform. By installing a room temperature sensor 13 on the user side, the room temperature on the user side can be detected and timely feedback can be given. The operating parameters can be adjusted in time according to the user's heating needs, so as to realize accurate heating in each heating area and ensure the room temperature, effectively improve the stable comfort of the room temperature, reduce heating energy consumption and costs, and maximize the heating effect.

[0150] The automatic operation and adjustment method of a heating system based on characteristic parameter identification comprises the following steps:

[0151] S1: List the process description equations of the primary side heat supply of the heat exchange station, the secondary side heat supply of the heat exchange station, the heat exchange of the heat exchanger, the heat dissipation of the user radiator, and the heat consumption of the user's enclosure structure, find out the characteristic constants that characterize the operating characteristics of the heating system, and convert the primary side water supply temperature T g_1 , Heat exchange station primary side return water temperature T h_1 , primary side flow rate F of heat exchange station g_1 , Heat exchange station secondary side water supply temperature T g_2 , Heat exchange station secondary side return water temperature T h_2, User indoor temperature T n and outdoor weather temperature T w The historical data of seven parameters are used as input, and the characteristic constants of the heating system are identified through the maximum likelihood estimation method, and the specific equations of each heat transfer process of the entire heating system are obtained;

[0152] The specific steps are as follows:

[0153] S11: Constructing the maximum likelihood estimation regression theory model:

[0154] 1) The maximum likelihood estimation model of the parameters of the two-variable linear regression model can be expressed as follows:

[0155] y=αx+β+ε (1)

[0156] In the formula: y is the response variable; x is the dependent variable; α is the slope of the line; β is the intercept of the line, and ε is the error;

[0157] The maximum likelihood estimation results of the model parameters are as follows:

[0158]

[0159] Where: is the sample mean of the response variable; is the sample mean of the dependent variable; is the maximum likelihood estimate of the slope of the straight line; is the maximum likelihood estimate of the line intercept;

[0160]

[0161] Where: x i is the dependent variable in the i-th group of samples; y i is the response variable in the i-th group of samples;

[0162]

[0163] Where: is the variance estimate; n is the sample size; ε i is the error in the i-th group of samples;

[0164] 2) The maximum likelihood estimation model of the parameters of the multivariate linear regression model can be expressed as follows:

[0165]

[0166] Where: y is the response variable; β0 is the spatial intercept; x1, x2, x k is the dependent variable related to the response variable; β1, β2, β k are the dependent variables x1, x2, x kThe corresponding weight factor; ε is the error;

[0167] For a model with n sets of measurements, the matrix expression is:

[0168] Y=Xβ+E (6)

[0169] In the formula: Y is the response variable sample matrix; X is the dependent variable sample matrix; β is the dependent variable weight factor matrix; E is the sample error matrix;

[0170]

[0171] The maximum likelihood estimation results of the model parameters are as follows:

[0172]

[0173] Where: is the maximum likelihood estimate of the dependent variable weight factor matrix;

[0174]

[0175] Where: is the variance estimate; n is the sample size;

[0176] S12: List the basic formulas for heat exchange stations and user heating:

[0177] The primary network heat supply can be expressed as follows:

[0178] Q g_1 =C1·(T g_1 -T h_1 )·F g_1 (13)

[0179] Where: Q g_1 is the heat supply of the primary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_1 , T h_1 are the primary side supply and return water temperatures of the heat exchange station, ℃; F g_1 is the water supply flow rate on the primary side of the heat exchange station, t / h;

[0180] The secondary network heat supply can be expressed as follows:

[0181] Q g_2 =C1·(T g_2 -T h_2 )·F g_2 (14)

[0182] Where: Q g_2 is the heat supply of the secondary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_2 , T h_2are the secondary side supply and return water temperatures of the heat exchange station, ℃; F g_2 is the water supply flow rate on the secondary side of the heat exchange station, t / h;

[0183] When the water on the primary and secondary sides of the heat exchanger flows in reverse, the heat exchange between the primary and secondary sides of the heat exchanger can be expressed as follows:

[0184]

[0185] Where: Q 1_2 is the heat exchange capacity of the primary and secondary sides of the heat exchange station, kw; C 1-2 is a static constant related to the heat transfer coefficient and heat transfer area of ​​the heat exchanger;

[0186] When the heat transfer coefficient of the radiator is a constant, the heat dissipation of the radiator can be expressed as follows:

[0187]

[0188] Where: Q s_s is the heat dissipation of the radiator, kw; C2 is a static constant related to the heat transfer coefficient and radiator area of ​​the radiator; T n is the user's indoor temperature, ℃;

[0189] The heat consumption of a building is mainly related to the heat transfer coefficient of the building envelope, the area of ​​the building envelope, the indoor temperature and the outdoor temperature. When the heat transfer coefficient of the building envelope is a constant, the heat consumption of the building can be expressed as follows:

[0190] Q j_h =C 3_1 ·[T n -(C 3_2 ·T w +C 3_3 )] (17)

[0191] Where: Q j_h is the heat consumption of the building envelope, kw; C 3_1 is a static constant related to the heat transfer coefficient and area of ​​the enclosure structure; T w is the outdoor temperature from meteorological data, ℃; 3_2 , C 3_3 is the correction constant for the outdoor temperature in the meteorological data;

[0192] S13: Count the primary side water supply temperature T of each heat exchange station g_1 , primary side return water temperature T h_1 , primary side flow F g_1 , secondary side water supply temperature T g_2 , secondary side return water temperature T h_2 , User indoor temperature T nHistorical operation data and the outdoor temperature forecast value T at the corresponding time w ;

[0193] S14: Identification of characteristic parameters of heating system:

[0194] 1) C1 is related to the specific heat capacity of water. According to the principle of dimensional harmony, C1 is 1.1625;

[0195] 2) Identify C by using the two-variable linear regression model in step S11 1_2 :

[0196] C 1_2 Identify the maximum likelihood estimation model:

[0197]

[0198] Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement;

[0199] 3) Identify C2 based on the two-variable linear regression model in step S11:

[0200] C2 identification maximum likelihood estimation model:

[0201]

[0202] Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement;

[0203] 4) Identify 1 / C by using the multivariate linear regression model in step S11 3_1 +1 / C2, C 3_2 , C 3_3 :

[0204] 1 / C 3_1 +1 / C2, C 3_2 , C 3_3 Identify the maximum likelihood estimation model:

[0205]

[0206] Where: ε is the random error caused by the system and measurement;

[0207] C 3_1 Based on the C2 identification result in step 3) and 1 / C 3_1 +1 / C2 identification results show that:

[0208]

[0209] S2: Based on the identification of characteristic parameters of the heating system, a heat load prediction model, a primary network flow distribution model, a heat source water supply temperature recommended value calculation model, a user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and a heat exchange station heat consumption evaluation model are constructed. The heat load of the entire heating system is predicted by the load prediction model, the primary network distribution flow of each heat exchange station is determined by the primary network flow distribution model, the heat source water supply temperature is determined by the heat source water supply temperature recommended value calculation model, the user indoor temperature average value within the heating area of ​​the heat exchange station is estimated by the user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and the heat consumption evaluation model of the heat exchange station is used to evaluate the heat consumption of the heat exchange station.

[0210] Based on the identification of characteristic parameters of the heating system, a heat load prediction model is constructed, which includes the following steps:

[0211] (1) Given the external environmental variables T of each heat exchange station w (n);

[0212] (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n);

[0213] (3) Predict the heat load Q0(n) of each heat exchange station:

[0214] The heat load prediction formula for each heat exchange station is as follows:

[0215] Q0(n)=C 3_1 (n)·{T n (n)-[C 3_2 (n)·T w (n)+C 3_3 (n)]} (22)

[0216] Where: Q0(n) is the predicted value of heat load of each heat exchange station, kw; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the meteorological data of each heat exchange station;

[0217] (4) Predict the heat load Q0 of the entire heating system:

[0218] The heat load of the entire heating system is the sum of the heat loads of each heat exchange station:

[0219] Q0=∑Q0(n) (23)

[0220] Based on the identification of characteristic parameters of the heating system, a primary network flow distribution model is constructed, which includes the following steps:

[0221] (1) Given the external environmental variables T of each heat exchange station at the same time w (n);

[0222] (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n);

[0223] (3) Calculate the total flow rate F of the primary network of the heat source g_0 ;

[0224] (4) Determine the primary network distribution flow F of each heat exchange station g_1 (n):

[0225] The calculation formula for the primary network flow of each heat exchange station is as follows:

[0226]

[0227] Where: C 3_1 is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and area of ​​the enclosure structure; C 3_2 , C 3_3 is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the relevant meteorological data; n and m represent heat exchange station n and heat exchange station m respectively.

[0228] Based on the identification of characteristic parameters of the heating system, a heat source water supply temperature recommendation value calculation model is constructed, including the following steps:

[0229] (1) Given the external environmental variables T of the heat exchange station w (n);

[0230] (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n);

[0231] (3) Set the expected value T of the indoor temperature in the heating area of ​​the heat exchange station n (n);

[0232] (4) Give the recommended value of heat source water supply temperature T g_0 :

[0233] The heat source water supply temperature is obtained by solving the following implicit function:

[0234]

[0235] Where: C1(n) is 1.1625; C2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and radiator area of ​​each heat exchange station; C 1_2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and heat exchange area of ​​each heat exchange station; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system based on the outdoor temperature in the meteorological data of each heat exchange station.

[0236] Based on the identification of characteristic parameters of the heating system, a model for estimating the average indoor temperature of users in the heating area of ​​the heat exchange station is constructed, including the following steps:

[0237] (1) Given the external environmental variables T of the heat exchange station w (n);

[0238] (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n);

[0239] (3) Given the heat source water supply temperature T g_0 ;

[0240] (4) By solving the implicit function formula (25) above, the average room temperature T of users in the heating area of ​​the heat exchange station is estimated. n (n):

[0241] Based on the identification of characteristic parameters of the heating system, a heat exchange station heat consumption evaluation model is constructed, which is divided into the following two cases:

[0242] 1) Heat consumption evaluation of different heat exchange stations in the same heating season includes the following steps:

[0243] (1) Statistics on the heating area F(n) and the cumulative annual heating supply of each heat exchange station The cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station;

[0244] (2) Calculate the annual heat supply per unit heating area of ​​each heat exchange station

[0245] The calculation formula for the annual heat supply per unit heating area is as follows:

[0246]

[0247] (3) Compare the heat supply per unit area of ​​each heat exchange station and evaluate the heat consumption;

[0248] 2) The heat consumption evaluation of the same heat exchange station in different heating seasons includes the following steps:

[0249] (1) Calculate the heating area F(n) and the cumulative annual heating supply of each heat exchange station and heating degree days HDD18, the cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station;

[0250] (2) Calculate the heat consumption per square meter of heating area per heating degree-day of each heat exchange station

[0251] The calculation formula of heat consumption per square meter of heating area per unit heating degree day is as follows:

[0252]

[0253] (3) Compare the heat consumption per square meter of heating area per heating degree-day of each heat exchange station and conduct heat consumption evaluation.

[0254] The implementation of the method of the present invention requires the following seven conditions:

[0255] 1) The heat exchange station needs to be transformed into an unmanned station, so that the heating parameters of the heat exchange station can be sensed and the heating equipment can be automatically controlled;

[0256] 2) Calculate the actual heating area F of each heat exchange station;

[0257] 3) Get the outdoor temperature T from the weather forecast w data;

[0258] 4) Measure and upload the heat source water supply flow F g_0 ;

[0259] 5) Measure and upload the primary side water supply temperature T of each heat exchange station in the heating system g_1 , primary side return water temperature T h_1 , primary side flow F g_1 , secondary side water supply temperature T g_2 , secondary side return water temperature T h_2 , secondary side flow F g_2 And the user's indoor temperature T n Operational data;

[0260] 6) The performance of the data acquisition system should be stable and reliable, and both the static and dynamic performance should meet the specified technical indicators;

[0261] 7) The system management platform computing software should be correct, reliable, testable, easy to use and maintain.

[0262] The process and principle of the method of the present invention:

[0263] 1. Technical feasibility

[0264] During the operation and management of the heating system, heating companies store a large amount of historical operation data. These data reflect the potential operating characteristics of the heating system, making it possible to use machine learning algorithms to identify the characteristic parameters of the heating system and, on this basis, establish heat load forecasting and automatic operation and regulation models for the heating network.

[0265] 2. Maximum Likelihood Estimation Analytical Algorithm Basics

[0266] Maximum likelihood estimation is a statistical method and a theoretical point estimation method. The basic idea is that when n groups of sample observations are randomly selected from the model population, the most reasonable parameter estimator should maximize the probability of extracting the n groups of sample observations from the model.

[0267] 1) Maximum Likelihood Estimation of Parameters of a Two-Variable Linear Regression Model

[0268] The model is as follows:

[0269] y=αx+β+ε (28)

[0270] Assuming that the model error follows a normal distribution, the following formula holds true:

[0271]

[0272] The likelihood function of the two-variable linear regression model is as follows:

[0273]

[0274] The log-likelihood function of formula (30) is:

[0275]

[0276] The first-order condition for maximization is:

[0277]

[0278] The maximum likelihood estimate is:

[0279]

[0280] 2) Maximum likelihood estimation of multivariate linear regression model parameters

[0281] The model is as follows:

[0282] y=β0+β1x1+β2x2+...+β k x k +ε (38)

[0283] For a model with n sets of measurements, the matrix expression is:

[0284] Y=Xβ+E (39)

[0285]

[0286] Assuming that the model error follows a normal distribution, the likelihood function is:

[0287]

[0288] The log-likelihood function is:

[0289]

[0290] The first-order condition for maximization is:

[0291]

[0292] The maximum likelihood estimate is:

[0293]

[0294]

[0295] 3. Description of heating system operation

[0296] Construct a heating system, including the primary side pipe network of the heat exchange station, the heat exchanger, the secondary side pipe network of the heat exchange station and the heat user. There are 8 operating variables involved in the heating system, which are:

[0297] 1) User's indoor temperature T n ;

[0298] 2) Outdoor temperature T w ;

[0299] 3) Primary side water supply temperature T of heat exchange station g_1 ;

[0300] 4) Heat exchange station primary side return water temperature T h_1 ;

[0301] 5) Primary side water supply flow F of heat exchange station g_1 ;

[0302] 6) Heat exchange station secondary side water supply temperature T g_2 ;

[0303] 7) Heat exchange station secondary side return water temperature T h_2 ;

[0304] 8) Secondary water supply flow F of heat exchange station g_2 .

[0305] When the heat loss of the heat exchanger and pipeline is ignored, the heat supply of the primary network is equal to the heat exchange of the heat exchanger:

[0306]

[0307] When the heat loss of heat exchanger and pipeline is ignored, the heat supply of primary network is equal to that of secondary network:

[0308] (T g_1 -T h_1 )·F g_1 =(T g_2 -T h_2 )·F g_2 (51)

[0309] Combining equations (50) and (51), we can get:

[0310]

[0311] When the heat loss of the heat exchanger and pipeline is ignored, the heat supply of the primary network is equal to the heat dissipation of the radiator:

[0312]

[0313] Combining equations (51) and (53), we can obtain:

[0314]

[0315]

[0316] When the heat loss of heat exchanger and pipeline is ignored, the heat supply of primary network is equal to the heat consumption of user's enclosure structure:

[0317] C1·(T g_1 -T h_1 )·F g_1 =C 3_1 ·[T n -(C 3_2 ·T w +C 3_3 )] (56)

[0318] Formula (56) is transformed into:

[0319]

[0320] Substituting equation (57) into equations (54) and (55), we can obtain:

[0321]

[0322] Substituting equations (58) and (59) into equation (52), we can obtain:

[0323]

[0324] From formula (56), we can get:

[0325]

[0326] Combining equation (60) and equation (61), we can get:

[0327]

[0328] The above formula (62) can be expressed as the following functional relationship:

[0329] f(T g_1 ,F g_1 ,F g_2 ,T n ,T w ,C1,C2,C 1_2 ,C 3_1 ,C 3_2 ,C 3_3 )=0 (63)

[0330] After the characteristic parameters of the heating system are identified, the average indoor temperature of users in the heating area of ​​the heat exchange station is T n It can be expressed as the primary side water supply temperature T of the heat exchange station g_1 , primary side water supply flow F of heat exchange station g_1 , Heat exchange station secondary side water supply flow F g_2 , outdoor temperature T w The functional relationship of the four variables is:

[0331] T n =g(T g_1 ,F g_1 ,F g_2 ,T w ) (64)

[0332] It can be seen from equations (58) and (59) that after the characteristic parameters of the heating system are identified, the secondary side water supply temperature T g_2 , Heat exchange station secondary side return water temperature T h_2 Only with the secondary side water supply flow F of the heat exchange station g_2 and indoor temperature T n , outdoor temperature T w The secondary side supply and return water temperature of the heat exchange station can be expressed as the following function:

[0333] T g_2 =u(F g_2 ,T n ,T w ) (65)

[0334] T h_2 =w(F g_2 ,T n ,T w ) (66)

[0335] It can be seen from formula (61) that after the static constant of the heating system is identified, the primary return water temperature T h_1 Only the primary side water supply temperature T g_1 , primary side water supply flow F of heat exchange station g_1 and indoor temperature T n , outdoor temperature T w The return water temperature on the primary side of the heat exchange station can be expressed as the following function:

[0336] T h_1 =χ(T g_1 ,F g_1 ,T n ,T w ) (67)

[0337] From formula (64), we can know that the user's indoor temperature T n It is about the primary side water supply temperature T of the heat exchange station g_1 , primary side water supply flow F of heat exchange station g_1 , Heat exchange station secondary side water supply flow F g_2 , outdoor temperature T w After substituting equation (64) into equation (67), the primary side return water temperature of the heat exchange station can be expressed as the following function:

[0338] T h_1 = z(T g_1 ,F g_1 ,F g_2 ,T w ) (68)

[0339] It can be seen from equations (64) to (66) and (68) that the primary side water supply temperature T g_1 , primary side water supply flow F of heat exchange station g_1 , Heat exchange station secondary side water supply flow F g_2 , outdoor temperature T w The four state parameters are defined as the basic state parameters of the heating system, and all other operating state parameters of the heating system can be uniquely represented by the four basic state parameters.

[0340] The conclusions of the heating system operation description provide a theoretical basis for identifying the number of input parameters required for the model.

[0341] When the automatic operation and regulation method of the heating system based on characteristic parameter identification of the present invention is applied, local weather forecast services are purchased to obtain weather data in real time, all data are stored in a data storage device, and the real-time data of each measuring point of the heating system is monitored in real time, and then the characteristic parameters of the heating system are identified according to the historical operation data of the heating system in the data storage device, and the heat load of the heating system is predicted according to the weather forecast, and a recommended value of the heat source water supply temperature is given. The PLC executes the adjustment instructions of the system management platform to control the primary-side electric regulating valves of each heat exchange station, and finally realizes "heating on demand" of the heat source and "distribution on demand" of the primary network flow of each heat exchange station, so as to achieve energy saving and environmental protection, and automatic operation and regulation of the heating system. The PLC controller automatically controls and regulates, so that the heat supply is adjusted according to user needs, which reduces the work intensity and operation difficulty of on-site operation and maintenance personnel, and reduces operation and management costs.

[0342] The present invention lists the process description equations of the primary side heat supply of the heat exchange station, the secondary side heat supply of the heat exchange station, the heat exchange of the heat exchanger, the heat dissipation of the user radiator, and the heat consumption of the user enclosure structure, and finds the characteristic constants that characterize the operating characteristics of the heating system. g_1 , Heat exchange station primary side return water temperature T h_1 , primary side flow rate F of heat exchange station g_1 , Heat exchange station secondary side water supply temperature T g_2 , Heat exchange station secondary side return water temperature T h_2 , User indoor temperature T n , outdoor weather temperature T w The historical data of seven parameters are used as input, and the characteristic constants of the heating system are identified through the maximum likelihood estimation method. The specific equations of each heat transfer process in the entire heating system are obtained. On this basis, the heat load prediction, heat source water supply temperature determination and primary network flow distribution are completed to realize the automatic operation and regulation of the heating system. Guided by the theoretical heat transfer formula, the prediction and control model is closer to the actual operation law of the heating system, which reduces the prediction error and further reduces the amount of sample data.

[0343] The prediction model in the heat load prediction process of the present invention can truly reflect the problems of heat transfer process between heat source, heat exchange station and heat users in the heating system; it can reduce the deviation between the theoretical calculated value of the heat source water supply temperature and the actual demand; it solves the problem of long adjustment time in the process of primary network flow distribution adjustment based solely on the return water temperature, and finally realizes "heating on demand" of the heat source and "distribution on demand" of the primary network flow of each heat exchange station, so as to achieve the purpose of energy saving, environmental protection and automatic operation and adjustment of the heating system.

[0344] The above contents are further detailed descriptions of the present invention in combination with the preferred technical solutions, and the specific implementation of the invention cannot be limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, simple deductions and substitutions can be made without departing from the concept of the present invention, which should be regarded as the protection scope of the present invention.

Claims

1. A method for automatic operation and adjustment of a heating system based on characteristic parameter identification, characterized in that: The following steps are involved: S1: List the process description equations of the primary side heat supply of the heat exchange station, the secondary side heat supply of the heat exchange station, the heat exchange of the heat exchanger, the heat dissipation of the user radiator, and the heat consumption of the user's enclosure structure, find out the characteristic constants that characterize the operating characteristics of the heating system, and convert the primary side water supply temperature T g_1 , Heat exchange station primary side return water temperature T h_1 , primary side flow rate of heat exchange station F g_1 , Heat exchange station secondary side water supply temperature T g_2 , Heat exchange station secondary side return water temperature T h_2 , User indoor temperature T n and outdoor weather temperature T w The historical data of seven parameters are used as input, and the characteristic constants of the heating system are identified by the maximum likelihood estimation method, and the specific equations of each heat transfer process of the entire heating system are obtained; the following steps are included: S11: Constructing the maximum likelihood estimation regression theory model: 1) The maximum likelihood estimation model of the parameters of the two-variable linear regression model can be expressed as follows: y=αx+β+ε (1) In the formula: y is the response variable; x is the dependent variable; α is the slope of the line; β is the intercept of the line, and ε is the error; The maximum likelihood estimation results of the model parameters are as follows: Where: is the sample mean of the response variable; is the sample mean of the dependent variable, is the maximum likelihood estimate of the slope of the straight line; is the maximum likelihood estimate of the line intercept; Where: x i is the dependent variable in the i-th group of samples; y i is the response variable in the i-th group of samples; Where: is the variance estimate; n is the sample size; ε i is the error in the i-th group of samples; 2) The maximum likelihood estimation model of the parameters of the multivariate linear regression model can be expressed as follows: y=β0+β1x1+β2x2+...+β k x k +e (5) Where: y is the response variable; β0 is the spatial intercept; x1, x2, x k is the dependent variable related to the response variable; β1, β2, β k are the dependent variables x1, x2, x k The corresponding weight factor; ε is the error; For a model with n sets of measurements, the matrix expression is: Y=Xβ+E (6) In the formula: Y is the response variable sample matrix; X is the dependent variable sample matrix; β is the dependent variable weight factor matrix; E is the sample error matrix; The maximum likelihood estimation results of the model parameters are as follows: Where: is the maximum likelihood estimate of the dependent variable weight factor matrix; Where: is the variance estimate; n is the sample size; S12: List the basic formulas for heat exchange stations and user heating: The primary network heat supply can be expressed as follows: Q g_1 =C1·(T g_1 -T h_1 )·F g_1 (13) Where: Q g_1 is the heat supply of the primary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_1 , T h_1 are the primary side supply and return water temperatures of the heat exchange station, ℃; F g_1 is the water supply flow rate on the primary side of the heat exchange station, t / h; The secondary network heat supply can be expressed as follows: Q g_2 =C1·(T g_2 -T h_2 )·F g_2 (14) Where: Q g_2 is the heat supply of the secondary network of the heat exchange station, KW; C1 is a static constant related to the specific heat capacity of water; T g_2 、T h_2 are the secondary side supply and return water temperatures of the heat exchange station, ℃; F g_2 is the water supply flow rate on the secondary side of the heat exchange station, t / h; When the water on the primary and secondary sides of the heat exchanger flows in reverse, the heat exchange between the primary and secondary sides of the heat exchanger can be expressed as follows: Where: Q 1_2 is the heat exchange capacity of the primary and secondary sides of the heat exchange station, kw; C 1-2 is a static constant related to the heat transfer coefficient and heat transfer area of ​​the heat exchanger; When the heat transfer coefficient of the radiator is a constant, the heat dissipation of the radiator can be expressed as follows: Where: Q s_s is the heat dissipation of the radiator, kw; C2 is a static constant related to the heat transfer coefficient and radiator area of ​​the radiator; T n is the user's indoor temperature, ℃; The heat consumption of a building is mainly related to the heat transfer coefficient of the building envelope, the area of ​​the building envelope, the indoor temperature and the outdoor temperature. When the heat transfer coefficient of the building envelope is a constant, the heat consumption of the building can be expressed as follows: Q j_h =C 3_1 ·[T n -(C 3_2 ·T w +C 3_3 )] (17) Where: Q j_h is the heat consumption of the building envelope, kw; C 3_1 is a static constant related to the heat transfer coefficient and area of ​​the enclosure structure; T w is the outdoor temperature from meteorological data, ℃; 3_2 , C 3_3 is the correction constant for the outdoor temperature in the meteorological data; S13: Count the primary side water supply temperature T of each heat exchange station g_1 , primary side return water temperature T h_1 , primary side flow F g_1 , secondary side water supply temperature T g_2 , secondary side return water temperature T h_2 , User indoor temperature T n Historical operation data and the outdoor temperature forecast value T at the corresponding time w ; S14: Identification of characteristic parameters of heating system: 1) C1 is related to the specific heat capacity of water. According to the principle of dimensional harmony, C1 is 1.1625; 2) Identify C by using the two-variable linear regression model in step S11 1_2 : C 1_2 Identify the maximum likelihood estimation model: Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement; 3) Identify C2 based on the two-variable linear regression model in step S11: C2 identification maximum likelihood estimation model: Where: β is the deviation value caused by the heat exchange loss of the heat exchanger and the heat dissipation loss of the pipeline in the actual process; ε is the random error caused by the system and measurement; 4) Identify 1 / C by using the multivariate linear regression model in step S11 3_1 +1 / C2, C 3_2 , C 3_3 : 1 / C 3_1 +1 / C2, C 3_2 , C 3_3 Identify the maximum likelihood estimation model: Where: ε is the random error caused by the system and measurement; C 3_1 Based on the C2 identification result in step 3) and 1 / C 3_1 +1 / C2 identification results show that: S2: Based on the identification of characteristic parameters of the heating system, a heat load prediction model, a primary network flow distribution model, a heat source water supply temperature recommended value calculation model, a user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and a heat exchange station heat consumption evaluation model are constructed. The heat load of the entire heating system is predicted by the load prediction model, the primary network distribution flow of each heat exchange station is determined by the primary network flow distribution model, the heat source water supply temperature is determined by the heat source water supply temperature recommended value calculation model, the user indoor temperature average value within the heating area of ​​the heat exchange station is estimated by the user indoor temperature average value estimation model within the heating area of ​​the heat exchange station, and the heat consumption evaluation model of the heat exchange station is used to evaluate the heat consumption of the heat exchange station.

2. The method for automatic operation and adjustment of a heating system based on characteristic parameter identification according to claim 1 is characterized in that: In step S2, a heat load prediction model is constructed based on the identification of characteristic parameters of the heating system, including the following steps: (1) Given the external environmental variables T of each heat exchange station w (n); (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n); (3) Predict the heat load Q0(n) of each heat exchange station: The heat load prediction formula for each heat exchange station is as follows: Q0(n)=C 3_1 (n)·{T n (n)-[C 3_2 (n)·T w (n)+C 3_3 (n)]} (22) Where: Q0(n) is the predicted value of heat load of each heat exchange station, kw; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the meteorological data of each heat exchange station; (4) Predict the heat load Q0 of the entire heating system: The heat load of the entire heating system is the sum of the heat loads of each heat exchange station: Q0=∑Q0(n) (23).

3. The automatic operation and adjustment method of a heating system based on characteristic parameter identification according to claim 1 is characterized in that: In step S2, a primary network flow distribution model is constructed based on the identification of characteristic parameters of the heating system, including the following steps: (1) Given the external environmental variables T of each heat exchange station at the same time w (n); (2) Set the expected value T of the indoor temperature in the heating area of ​​each heat exchange station n (n); (3) Calculate the total flow rate F of the primary network of the heat source g_0 ; (4) Determine the primary network distribution flow F of each heat exchange station g_1 (n): The calculation formula for the primary network flow of each heat exchange station is as follows: Where: C 3_1 is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and area of ​​the enclosure structure; C 3_2 , C 3_3 is the corrected identification value of the characteristic parameter of the heating system of the outdoor temperature in the relevant meteorological data; n and m represent heat exchange station n and heat exchange station m respectively.

4. The automatic operation and adjustment method of a heating system based on characteristic parameter identification according to claim 1 is characterized in that: In step S2, a heat source water supply temperature recommendation value calculation model is constructed based on the identification of characteristic parameters of the heating system, including the following steps: (1) Given the external environmental variables T of the heat exchange station w (n); (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n); (3) Set the expected value T of the indoor temperature in the heating area of ​​the heat exchange station n (n); (4) Give the recommended value of heat source water supply temperature T g_0 : The heat source water supply temperature is obtained by solving the following implicit function: Where: C1(n) is 1.1625; C2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and radiator area of ​​each heat exchange station; C 1_2 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and heat exchange area of ​​each heat exchange station; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system based on the outdoor temperature in the meteorological data of each heat exchange station.

5. The automatic operation and adjustment method of a heating system based on characteristic parameter identification according to claim 1 is characterized in that: In step S2, a model for estimating the average indoor temperature of users in the heating area of ​​the heat exchange station is constructed based on the identification of characteristic parameters of the heating system, including the following steps: (1) Given the external environmental variables T of the heat exchange station w (n); (2) Given the primary and secondary network water supply flow variables F of the heat exchange station g_1 (n),F g_2 (n); (3) Given the heat source water supply temperature T g_0 ; (4) Estimate the average room temperature T of users in the heating area of ​​the heat exchange station by solving the following implicit function formula (25) n (n): Where: C1(n) is 1.1625; C2(n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and radiator area of ​​each heat exchange station; C 1_2 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient and heat exchange area of ​​each heat exchange station; C 3_1 (n) is the identification value of the characteristic parameters of the heating system related to the heat transfer coefficient of the enclosure structure and the area of ​​the enclosure structure of each heat exchange station; C 3_2 (n), C 3_3 (n) is the corrected identification value of the characteristic parameter of the heating system based on the outdoor temperature in the meteorological data of each heat exchange station.

6. The automatic operation and adjustment method of a heating system based on characteristic parameter identification according to claim 1 is characterized in that: In step S2, a heat exchange station heat consumption evaluation model is constructed based on the identification of characteristic parameters of the heating system, which is divided into the following two cases: 1) Heat consumption evaluation of different heat exchange stations in the same heating season includes the following steps: (1) Statistics on the heating area F(n) and the cumulative annual heating supply of each heat exchange station The cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station; (2) Calculate the annual heat supply per unit heating area of ​​each heat exchange station The calculation formula for the annual heat supply per unit heating area is as follows: (3) Compare the heat supply per unit area of ​​each heat exchange station and evaluate the heat consumption; 2) The heat consumption evaluation of the same heat exchange station in different heating seasons includes the following steps: (1) Calculate the heating area F(n) and the cumulative annual heating supply of each heat exchange station and heating degree days HDD18, the cumulative annual heat supply is obtained by integrating the primary side flow meter and the primary side supply and return water temperature in the heat exchange station; (2) Calculate the heat consumption per square meter of heating area per heating degree-day of each heat exchange station The calculation formula of heat consumption per square meter of heating area per unit heating degree day is as follows: (3) Compare the heat consumption per square meter of heating area per heating degree-day of each heat exchange station and conduct heat consumption evaluation.

Images