Heat supply and heat exchange station optimization control method and system based on economic model predictive control

Abstract

The invention discloses a heating heat exchange station optimization control method and system based on economic model predictive control. The method comprises the following steps: S1, collecting historical operation data and meteorological data of a heat exchange station, and constructing a training data set to establish a dynamic mathematical model of the heat exchange station and identify model parameters; and S2, based on the dynamic mathematical model of the heat exchange station established in the step S1, taking the primary or secondary network water supply flow as a control variable, considering the heat supply demand and operation constraint of a heat user, minimizing a heat exchange station operation cost objective function, designing an economic model prediction controller, and reasonably adjusting the operation condition of the heat exchange station in time. According to the method, accurate prediction of key heat supply parameters such as water supply and return temperature and pressure can be achieved, then optimization regulation and control of the heat exchange station are achieved, meanwhile, the optimal economic performance can be guaranteed on the premise that the heat supply requirement of a heat user and safe operation are met, and therefore the heat supply cost is reduced, and energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal energy engineering, and particularly relates to an optimized control method and system for a heating heat exchange station based on economic model predictive control. Background Art

[0002] The urban central heating system is an important part of the energy system in northern cities of China, which is related to the vital interests of the people. This system mainly includes heat sources, primary pipe networks, heat exchange stations, secondary pipe networks, and heat users, etc. Among them, the heat exchange station is not only the link connecting the primary pipe network and the secondary pipe network, but also the key node for heat energy regulation in the entire heating system, which is directly related to the heating quality of heat users and the economic benefits of heating enterprises.

[0003] However, the existing central heating system is a complex distributed large system, which has obvious characteristics such as large lag, time-varying, non-linear, and strong coupling. In particular, due to the large thermal inertia of heat sources, heating pipe networks, and buildings, the change of meteorological parameters and parameters such as supply water temperature and supply water flow rate will have a long lag time on the room temperature of heat users. These characteristics bring many challenges to the regulation of heat exchange stations.

[0004] Currently, the regulation of heat exchange stations usually adjusts the heating parameters of heat exchange stations according to the operation experience of thermal management personnel or according to the complaint rate of heat users. However, this regulation method mainly has the following problems: 1) There is an obvious time delay between the start of the regulation action of the heat exchange station and the response of the indoor temperature of heat users. The delay time is different for different heat exchange stations and different weather conditions, working conditions, and secondary network structures. It is difficult to obtain the heating delay time under different working conditions online, resulting in inappropriate regulation time (too early or too late) and low user thermal comfort in the regulation of heat exchange stations; 2) The regulation of heat exchange stations mainly takes parameters such as the supply water temperature of the secondary pipe network and the average supply and return water temperature as the adjustment targets, while the supply and return water temperatures cannot fully reflect the actual heat demand of heat users; 3) The regulation of heat exchange stations is a non-linear control problem involving multiple variables, and the relationship between regulation variables and response characteristics is complex. The regulation method relying on subjective experience often requires a repeated calibration process of adjustment, stabilization, and re-adjustment for each station, with low efficiency and high requirements for the technical level of operators; 4) In actual regulation, on the one hand, it relies heavily on the experience and technical level of regulation personnel. On the other hand, the number of heat stations that need to be monitored and regulated by regulation personnel is large, and the workload of manual regulation is large. Even experienced regulation personnel are difficult to track and regulate each station in real time. Under such objective circumstances, regulation personnel mostly conduct unified and coarse-grained regulation when there is an obvious temperature drop or rise, or when there are many user complaints; 5) In actual regulation, in order to reduce the user complaint rate, regulation personnel often increase the heating supply, resulting in overheating of some systems and heat waste. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the technical problem to be solved by the present invention is to provide an optimized control method and system for a heating heat exchange station based on economic model predictive control. The method and system can accurately predict key heating parameters such as supply and return water temperatures and pressures, and then realize the optimized regulation of the heat exchange station. At the same time, on the premise of meeting the heating demands of heat users and safe operation, the best economic performance can be guaranteed, thereby reducing the heating cost and saving energy consumption.

[0006] To achieve the above object, on the one hand, the present invention discloses an optimized control method for a heat exchange station based on economic model predictive control, including the following steps:

[0007] S1: Collect the historical operation data and meteorological data of the heat exchange station, construct a training data set, establish a dynamic mathematical model of the heat exchange station and identify the model parameters, so as to realize the online prediction of key heating parameters;

[0008] S2: Based on the dynamic mathematical model of the heat exchange station established in step S1, taking the supply water flow of the primary or secondary network as the control variable, considering the heating demands of heat users and operation constraints, minimize the operation cost objective function of the heat exchange station, design an economic model predictive controller, and adjust the operation conditions of the heat exchange station in a timely and reasonable manner.

[0009] Preferably, the historical operation data of the heat exchange station in step S1 includes: the primary network supply water temperature T 1g (k), in °C; the primary network return water temperature T 1h (k), in °C; the primary network supply water pressure P 1g (k), in Mpa; the primary network return water pressure P 1h (k), in Mpa; the primary network supply water flow G1(k), in t / h; the secondary network supply water temperature T 2g (k), in °C; the secondary network return water temperature T 2h (k), in °C; the secondary network supply water pressure P 2g (k), in Mpa; the secondary network return water pressure P 2h (k), in Mpa; the secondary network supply water flow G2(k), in t / h; the meteorological data includes the outdoor temperature T out (k), in °C.

[0010] Preferably, in step S1, the method for establishing the dynamic mathematical model of the heat exchange station and identifying the model parameters includes the following steps:

[0011] S1.1: At the sampling times k = 1, 2..., K, collect the current operation data collected by the heat exchange station during the heating season: the primary network supply water temperature T 1g(k), in °C; the return water temperature T of the primary network 1h (k), in °C; the supply water pressure P of the primary network 1g (k), in Mpa; the return water pressure P of the primary network 1h (k), in Mpa; the supply water flow rate G1(k) of the primary network, in t / h; the supply water temperature T of the secondary network 2g (k), in °C; the return water temperature T of the secondary network 2h (k), in °C; the supply water pressure P of the secondary network 2g (k), in Mpa; the return water pressure P of the secondary network 2h (k), in Mpa; the supply water flow rate G2(k) of the secondary network, in t / h; and the outdoor temperature T at the corresponding moment out (k), in °C;

[0012] Define the column vector u(k) = [T 1g (k); P 1g (k); G1(k); G2(k); T out (k)] ∈ i 5 Formula (1.1);

[0013] Formula (1.1) is the input variable of the dynamic mathematical model of the heat exchange station; j 5 represents a 5th-order column vector.

[0014] The column vector y(k) = [T 1h (k); P 1h (k); ΔT2(k); ΔP2(k)] ∈ R 4 Formula (1.2);

[0015] Formula (1.2) is the output variable of the dynamic mathematical model of the heat exchange station; R 4 represents a 4th-order column vector.

[0016] Among them, VT2(k) = T 2g (k) - T 2h (k) Formula (1.3), representing the temperature difference between the supply and return water of the secondary network, in °C; VP2(k) = P 2g (k) - P 2h (k) Formula (1.4), representing the pressure difference between the supply and return water of the secondary network, in Mpa;

[0017] Again, based on Formulas (1.1) to (1.4), obtain the input-output data {u(k), y(k)} of K groups at the sampling moment k, and define the output matrix Y = [y(1), y(2), …, y(K)] ∈ R 4×K Formula (1.5);

[0018] S1.2: Recursively obtain the state variable x(k) according to formula (1.6):

[0019] x(k) = tanh(W in ×u(k) + W × x(k - 1)), k = 1, 2, … K Formula (1.6);

[0020] where x(k) ∈ j n , n is the dimension of the state variable, is an adjustable parameter, and the initial state x(0) = 0 is defined; the function tanh(·) is the hyperbolic tangent function; W in ∈ j n×5 is the input coefficient matrix, W ∈ j n×n is the state transition coefficient matrix (both W in and W are randomly generated within [-1, 1]); define the state matrix X = [x(1), x(2), …, x(K)] ∈ R n×K Formula (1.7);

[0021] S1.3: Calculate the output coefficient matrix between the state variable x(k) and the output variable y(k) according to formula (1.8)

[0022] W out = inv(X × X T + β × I n ) × X × Y T Formula (1.8);

[0023] In formula (1.8), inv(·) represents the matrix inversion operation, the superscript T represents the vector transpose operation, β is an adjustable weighting coefficient, represents the identity matrix of size n;

[0024] S1.4: Save the input coefficient matrix W in , the state transition coefficient matrix W, and W of formula (1.8) out .

[0025] Preferably, in formula (1.6), 500 ≤ n ≤ 1000, both W in and W are randomly generated within [-1, 1]; in formula (1.8), 10 -6 ≤ β ≤ 10 -2 .

[0026] Preferably, in step S2, the method for designing an economic model predictive controller includes the following steps:

[0027] S2.1: Set the prediction step size as P, the control step size as M, and satisfy M ≤ P. At the operation time l of the heat exchange station, denote M control sequences of the primary network water supply flow to be optimized and calculated as G1(l) = [G1(l + 1), G1(l + 2), …, G1(l + M)] Formula (2.1);

[0028] Denote M control sequences of the secondary network water supply flow to be optimized and calculated as G2(l) = [G2(l + 1), G2(l + 2), …, G2(l + M)] Formula (2.2); Then, based on Step S1.4, save the coefficient matrices W in , W, and W out , and recursively calculate the primary network return water temperature T 1h (l + i|l), the primary network return water temperature P 1h (l + i|l), the secondary network supply - return water temperature difference △T2(l + i|l), and the secondary network supply - return water pressure difference △P2(l + i|l) in the future time domain [l + 1, l + P], where i = 1, 2, …, P;

[0029]

[0030] Among them, T out (l + i) represents the predicted value of the outdoor temperature in the future time domain [l + 1, l + P], obtained from the weather forecast;

[0031]

[0032] Among them, ΔT2 is the secondary network supply - return water temperature difference, and ΔP2 is the secondary network supply - return water pressure difference.

[0033] S2.2: Define the heat exchange station operation cost objective function J1 and the control increment objective function J2 respectively:

[0034]

[0035]

[0036] In the objective function J1, c w is the specific heat capacity of water, J / (kg·℃); η1 and η2 correspond to the heat energy purchase price coefficient and the circulating pump power consumption price coefficient respectively. And when i > M, set G1(l + i) = G1(l + M) and G2(l + i) = G2(l + M). In the objective function J2, w1 and w2 correspond to the weighting coefficients of the change in the primary and secondary network water supply flows respectively;

[0037] S2.3: Construct the following constrained optimization problem:

[0038]

[0039]

[0040] In formula (2.8), Q is the user's expected heat load in the future time domain [l+1, l+P], with the unit of KW; △P 1set is the lower limit of the differential pressure between the supply and return water of the primary heat supply pipe network to ensure safe operation; △P 2set is the lower limit of the differential pressure between the supply and return water of the secondary heat supply pipe network to ensure safe operation.

[0041] S2.4: First, use the sequential quadratic programming method to solve the optimization problem described in step S2.3, and simultaneously obtain the optimal control sequences G1(l) and G2(l), and take the first terms of the optimal control sequences G1(l) and G2(l) respectively, that is, G1(l+1) and G2(l+1); Secondly, it is transmitted to the PLC control module of the heat exchange station through the network by the background processor for execution; Then, transfer to the l+1 moment, and repeat steps S2.1 to S2.4 to optimize and obtain the water supply flow of the primary network and the water supply flow of the secondary network, and execute the regulation of the heat exchange station.

[0042] On the other hand, the present invention discloses a control system for executing the above control method, including a sensor group, a data storage module, an analysis and processing module, and a control module;

[0043] The analysis and processing module includes a background processor;

[0044] The control module includes the PLC control module of the heat exchange station; the PLC control module is connected to the background processor;

[0045] The analysis and processing module is connected to the control module to transmit the obtained optimized heating parameters, and the control module is connected to the heating parameter control device of the heat exchange station to regulate the heating parameters of the heat exchange station according to the received optimized heating parameters;

[0046] The sensor group is arranged on the water pipe for monitoring and collecting the operation data of the heat exchange station;

[0047] The data storage module, connected to the sensor group, is used for collecting and storing the real-time and historical operation data of the heat exchange station and the meteorological data, and transmitting the operation data and the meteorological data to the analysis and processing module;

[0048] The analysis and processing module runs the control method according to the obtained data to calculate the optimized heating parameters.

[0049] Preferably, the heating parameters include the water supply flow of the primary network and the water supply flow of the secondary network.

[0050] Preferably, the heating parameter control device includes an electronic regulating valve, a frequency converter, and a circulating pump unit. The electronic regulating valve is arranged on the water pipe connecting the heat exchanger, the primary network, and the secondary network for regulating the water flow.

[0051] Preferably, the sensor group includes a pressure sensor, a flow sensor, and a temperature sensor.

[0052] Preferably, the water pipe includes a primary water supply pipe, a primary water return pipe, a secondary water supply pipe, and a secondary water return pipe;

[0053] The temperature sensor, the pressure sensor, and the regulating valve are sequentially arranged on the primary water supply pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the secondary water return pipe; the temperature sensor, the pressure sensor, and the valve are sequentially arranged on the secondary water supply pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the primary water return pipe.

[0054] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The method and system can accurately predict key heating parameters such as the temperature and pressure of the supply and return water, and then realize the optimized control of the heat exchange station. At the same time, on the premise of meeting the heating needs of heat users and ensuring safe operation, the best economic performance can be guaranteed, thereby reducing the heating cost and saving energy consumption. As specific effects: 1) The system and method of this patent establish a dynamic model based on weather forecast information and the heat exchange station system, which can realize the predictive control of the heat station, timely and reasonably adjust the operating conditions of the heat exchange station, reduce the lag time of the heat network, and ensure the heating quality. 2) The system and method of this patent take into account both the heat demand of users and the requirement of reducing the cost of heating operation, and comprehensively use two control means of qualitative regulation and quantitative regulation to achieve the purpose of energy conservation and consumption reduction. 3) The system and method of this patent reduce the dependence on the control personnel, realize the closed-loop control of "accurate heating" and "heating on demand", and create conditions for realizing an unmanned heat exchange station. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0056] Figure 1 It is a schematic structural diagram of the control system for this embodiment;

[0057] Figure 2 It is a schematic structural diagram of the dynamic mathematical model of the heat exchange station for this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0058] Hereinafter, the present invention will be specifically described by way of exemplary embodiments. However, it should be understood that, without further elaboration, the structures and features in one embodiment can also be beneficially combined with those in other embodiments.

[0059] It should be noted that in the description of the present invention, the indicated orientation or positional relationship is based on the positional relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the structure referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0060] As Figure 1 and Figure 2 shown, an optimized control method for a heat exchange station based on economic model predictive control includes the following steps:

[0061] S1: Collect the historical operation data and meteorological information of the heat exchange station, construct a training data set, establish a dynamic mathematical model of the heat exchange station and identify the model parameters, so as to realize the online prediction of heating parameters;

[0062] S2: Based on the dynamic mathematical model of the heat exchange station established in step S1, taking the water supply flow rate of the primary or secondary network as the control variable, considering the heating demand of heat users and operation constraints, minimize the operation cost objective function of the heat exchange station, design an economic model predictive controller, and adjust the operation conditions of the heat exchange station in a timely and reasonable manner.

[0063] As Figure 1 shown, the central heating system mainly includes a heat source 1, a primary pipe network 2, a heat exchanger 3, a secondary pipe network 4, and heat users 5, etc. The heat source 1 is connected to the heat exchanger 3 through the primary pipe network 2; the heat exchanger 3 is connected to the heat users 5 through the secondary pipe network 4.

[0064] The control system adopting the above control method includes a sensor group 6, a data storage module 7, an analysis and processing module 8, and a control module 9; the sensor group 6 can be set on the water pipes of the primary pipe network 2 and the secondary pipe network 4 for monitoring and collecting the operation data of the heat exchange station; the control module 9 is connected to the data storage module 7; the control module 9 is preferably a PLC.

[0065] The data storage module 7 is connected to the sensor group 6 and the meteorological station 10, and is used for collecting and storing the real-time and historical operation data of the heat exchange station and the real-time and historical meteorological information data, and transmitting the corresponding data to the analysis and processing module 8. The analysis and processing module 8 can be a processor such as a background server, and is connected through a network.

[0066] According to the obtained data, run the control method to calculate the optimized heating parameters. The analysis and processing module 8 is connected to the control module 9 to transmit the obtained optimized heating parameters. The control module 9 is connected to the heating parameter control device of the heat exchange station to adjust the heating parameters of the heat exchange station according to the received optimized heating parameters.

[0067] The control method and system of this embodiment establish a dynamic model based on meteorological forecast information and the heat exchange station system, which can accurately predict key heating parameters such as supply and return water temperatures and pressures, and then adjust the operating conditions of the heat exchange station in a timely and reasonable manner, reduce the lag time of the heat network, and ensure the heating quality. This method takes into account the two requirements of users' heat demand and cost reduction in heat supply operation, and comprehensively uses two control means of quality regulation and quantity regulation to achieve the purpose of energy conservation and consumption reduction. This method reduces the dependence on control personnel, realizes the closed-loop control of "precision heating" and "heat supply on demand", and creates conditions for the realization of unmanned heat exchange stations.

[0068] Among them, the heating parameters include the primary network supply water flow rate and the secondary network supply water flow rate.

[0069] Among them, the heating parameter control device includes an electronic control valve 11 for adjusting the water flow rate, a frequency converter 12, and a circulating pump unit, which are arranged on the water pipes connecting the heat exchanger 3, the primary pipe network 2, and the secondary pipe network 4. Specifically, the circulating pump unit includes a primary network circulating water pump 13 installed in the primary pipe network 2 and a secondary network circulating water pump 14 installed in the secondary pipe network 4.

[0070] Among them, the sensor group 6 includes a flow sensor, a temperature sensor, and a pressure sensor.

[0071] Specifically, the sensor group 6 includes a supply water temperature sensor 61, a supply water pressure sensor 62, a return water temperature sensor 63, and a return water pressure sensor 64. Specifically, the water pipes include a primary outlet pipe, a primary return pipe, a secondary outlet pipe, and a secondary return pipe; the temperature sensor, the pressure sensor, and the regulating valve are sequentially arranged on the primary outlet pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the secondary return pipe; the temperature sensor, the pressure sensor, and the valve are sequentially arranged on the secondary outlet pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the primary return pipe.

[0072] Among them, the historical operation data of the heat exchange station in step S1 include: the primary network supply water temperature T 1g (k), with the unit of °C; the primary network return water temperature T 1h (k), with the unit of °C; the primary network supply water pressure P 1g (k), with the unit of Mpa; the primary network return water pressure P 1h (k), with the unit of Mpa; the primary network supply water flow rate G1(k), with the unit of t / h; the secondary network supply water temperature T 2g (k), with the unit of °C; the secondary network return water temperature T 2h (k), with the unit of °C; the secondary network supply water pressure P 2g (k), with the unit of Mpa; the secondary network return water pressure P 2h(k), with the unit of Mpa; the secondary network water supply flow rate G2(k), with the unit of t / h; the meteorological data includes the outdoor temperature T at the corresponding moment out (k), with the unit of °C.

[0073] Specifically, in step S1, the method for establishing the dynamic mathematical model of the heat exchange station and identifying the model parameters includes the following steps:

[0074] S1.1: At the sampling moments k = 1, 2…, K, collect the operation data of the heat exchange station during the heating season at that time: the primary network water supply temperature T 1g (k), with the unit of °C; the primary network return water temperature T 1h (k), with the unit of °C; the primary network water supply pressure P 1g (k), with the unit of Mpa; the primary network return water pressure P 1h (k), with the unit of Mpa; the primary network water supply flow rate G1(k), with the unit of t / h; the secondary network water supply temperature T 2g (k), with the unit of °C; the secondary network return water temperature T 2h (k), with the unit of °C; the secondary network water supply pressure P 2g (k), with the unit of Mpa; the secondary network return water pressure P 2h (k), with the unit of Mpa; the secondary network water supply flow rate G2(k), with the unit of t / h; and the outdoor temperature T at the corresponding moment out (k), with the unit of °C;

[0075] Define the column vector u(k) = [T 1g (k); P 1g (k); G1(k); G2(k); T out (k)] ∈ j 5 Formula (1.1);

[0076] j 5 represents a 5th-order column vector

[0077] as the input variable of the dynamic mathematical model of the heat exchange station; the column vector y(k) = [T 1h (k); P 1h (k); ΔT2(k); ΔP2(k)] ∈ R 4 Formula (1.2); R 4 represents a 4th-order column vector as the output variable of the dynamic mathematical model of the heat exchange station;

[0079] Among them, VT2(k) = T 2g (k) - T 2h (k) Formula (1.3), representing the temperature difference between the supply and return water of the secondary network, with the unit of °C; VP2(k) = P 2g (k) - P 2h(k) Formula (1.4), representing the pressure difference between the supply and return water of the two networks, with the unit of Mpa;

[0080] Thirdly, based on Formulas (1.1) to (1.4), K groups of input-output data {u(k), y(k)} at the sampling time k are obtained, and the output matrix Y = [y(1), y(2), …, y(K)] ∈ R 4×K Formula (1.5);

[0081] S1.2: First, the state variable x(k) is recursively obtained according to Formula (1.6):

[0082] x(k) = tanh(W in ×u(k) + W×x(k - 1)), k = 1, 2, ……, K Formula (1.6);

[0083] In Formula (1.6), n is the dimension of the state variable, which is an adjustable parameter, and the initial state x(0) = 0 is defined; the function tanh(·) is the hyperbolic tangent function, is the input coefficient matrix, is the state transition coefficient matrix (both W in and W are randomly generated within [-1, 1]);

[0084] Then, the state matrix X = [x(1), x(2),..., x(K)] ∈ R n×K Formula (1.7);

[0085] S1.3: Calculate the output coefficient matrix between the state variable x(k) and the output variable y(k) according to Formula (1.8)

[0086] W out = inv(X×X T + β×I n )×X×Y T Formula (1.8);

[0087] In Formula (1.8), inv(·) represents the matrix inversion operation, the superscript T represents the vector transpose operation, β is an adjustable weighting coefficient, represents the identity matrix of size n;

[0088] S1.4: Save the input coefficient matrix W in 、the state transition coefficient matrix W and W of Formula (1.8) out .

[0089] In step S2, the method for designing the economic model predictive controller includes the following steps:

[0090] S2.1: First, set the prediction step size as P, the control step size as M, and satisfy M ≤ P;

[0091] Secondly, at the operation time l of the heat exchange station, M control sequences of the primary network water supply flow to be optimized and calculated are

[0092] G1(l) = [G1(l + 1), G1(l + 2), ……, G1(l + M)] Formula (2.1);

[0093] M control sequences of the secondary network water supply flow to be optimized and calculated are

[0094] G2(l) = [G2(l + 1), G2(l + 2), ……, G2(l + M)] Formula (2.2);

[0095] Then, based on the coefficient matrices W in , W and W out saved in step S1.4, recursively calculate the primary network return water temperature T 1h (l + i|l), the primary network return water temperature P 1h (l + i|l), the secondary network supply - return water temperature difference △T2(l + i|l), and the secondary network supply - return water pressure difference △P2(l + i|l) in the future time domain [l + 1, l + P], where i = 1, 2, ……, P;

[0096]

[0097] Among them, T out (l + i) represents the predicted value of the outdoor temperature in the future time domain [l + 1, l + P], obtained from the weather forecast;

[0098]

[0099] Among them, ΔT2 is the secondary network supply - return water temperature difference, and ΔP2 is the secondary network supply - return water pressure difference.

[0100] S2.2: Define the operation cost objective function J1 and the control increment objective function J2 of the heat exchange station respectively:

[0101]

[0102]

[0103] Among them, in the objective function J1, c w is the specific heat capacity of water, J / (kg·℃); η1 and η2 correspond to the heat energy purchase price coefficient and the circulating pump power consumption price coefficient respectively, and when i > M, set G1(l + i) = G1(l + M) and G2(l + i) = G2(l + M);

[0104] In the objective function J2, w1 and w2 are the weighted coefficients corresponding to the changes in the water supply flow rates of the primary and secondary water supply networks respectively;

[0105] S2.3: Construct the following constrained optimization problem:

[0106]

[0107]

[0108] In formula (2.8), Q is the expected heat load of users in the time domain of [l + 1, l + P] in the future, with the unit of KW; △P 1set is the lower limit value of the pressure difference between the supply and return water of the primary heat supply network to ensure safe operation; △P 2set is the lower limit value of the pressure difference between the supply and return water of the secondary heat supply network to ensure safe operation.

[0109] S2.4: First, use the sequential quadratic programming method to solve the optimization problem described in step S2.3, and simultaneously obtain the optimal control sequences G1(l) and G2(l), and take the first terms of the optimal control sequences G1(l) and G2(l) respectively, that is, G1(l + 1) and G2(l + 1); secondly, transfer them through the background processor to the PLC control module of the heat exchange station via the network for execution; then, transfer to the (l + 1)-th moment, and repeat steps S2.1 to S2.4 to optimize the water supply flow rate of the primary network and the water supply flow rate of the secondary network, and perform the regulation of the heat exchange station.

[0110] Specifically, in formula (1), 500 ≤ n ≤ 1000, and both W in and W are randomly generated within [-1, 1]; in formula (2), 10 -6 ≤ β ≤ 10 -2 .

[0111] The above are only the preferred embodiments of the present invention. The conventional technologies in the field are not described in detail in the embodiments. The above embodiments do not limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An optimized control method for a heat exchange station based on economic model predictive control, characterized in that: It includes the following steps: S1: First, collect the historical operation data of the heat exchange station and the meteorological data corresponding to the sampling time of the operation data; then, store the historical operation data as historical data and meteorological data, construct a training data set to establish a dynamic mathematical model of the heat exchange station and identify the model parameters, so as to realize the online prediction of heating parameters; S2: Based on the dynamic mathematical model of the heat exchange station established in step S1, with the primary or secondary network water supply flow rate as the control variable, determine the objective function of minimizing the operation cost of the heat exchange station, design an economic model predictive controller, and adjust the operation conditions of the heat exchange station.

2. The optimized control method for a heat exchange station based on economic model predictive control according to claim 1, characterized in that: In step S1, the operation data includes: The temperature T of the water supply for one network 1g (k), with the unit of °C; Return water temperature T of a network 1h (k), unit: °C; The water supply pressure P of a single network 1g (k), with the unit of Mpa; The pressure P of the return water in a network 1h (k), with the unit of Mpa; The primary network water supply flow rate G1(k), with the unit of t / h; Two-network water supply temperature T 2g (k), in °C; The return water temperature T of the secondary network 2h (k), in units of °C; Two-network water supply pressure P 2g (k), with the unit of Mpa; Backwater pressure P of the secondary network 2h (k), with the unit of Mpa; The secondary network water supply flow rate G2(k), with the unit of t / h; The meteorological data includes the outdoor temperature T out (k) at the corresponding moment of the operation data, with the unit of °C.

3. The optimized control method for a heat exchange station based on economic model predictive control according to claim 2, characterized in that: In step S1, the method for establishing a dynamic mathematical model of the heat exchange station and identifying the model parameters includes the following steps: S1.1: First, at sampling times k = 1, 2, …, K, collect the operation data collected by the heat exchange station during the heating season and the outdoor temperature T out (k) at the corresponding time; Then, establish a definition column vector for the input variables of the dynamic mathematical model of the heat exchange station; u(k) = [T 1g (k); P 1g (k); G1(k); G2(k); T out (k)] ∈ j 5 Equation (1.1); where j 5 represents a 5th-order column vector; Secondly, establish a column vector for the output variables of the dynamic mathematical model of the heat exchange station; y(k) = [T 1h (k); P 1h (k); ΔT2(k); ΔP2(k)] ∈ R 4 Equation (1.2); wherein, R 4 represents a 4th-order column vector; Determine the temperature difference between the supply and return water of the secondary network, with the unit of °C; VT2(k) = T 2g (k) - T 2h (k) Equation (1.3); Determine the pressure difference between the supply and return water of the secondary network, with the unit of Mpa; VP2(k) = P 2g (k) - P 2h (k) Equation (1.4); Again, based on formulas (1.1) to (1.4), obtain K groups of input-output data {u(k), y(k)} at the sampling time k, and define the output matrix; Y = [y(1), y(2), …, y(K)] ∈ R 4×K Formula (1.5); S1.2: First, recursively obtain the state variable x(k) according to formula (1.6): x(k) = tanh(W in ×u(k) + W×x(k - 1)), k = 1, 2, …K Equation (1.6); In formula (1.6), x(k) ∈ j n , where n is the dimension of the state variable, is an adjustable parameter, and the initial state x(0) = 0 is defined; the function tanh(·) is the hyperbolic tangent function; W in ∈ j n×5 is the input coefficient matrix, and W ∈ j n×n is the state transition coefficient matrix; W in and W are both randomly generated within [-1, 1]; Then, define the state matrix; X = [x(1), x(2), …, x(K)] ∈ R n×K Formula (1.7); S1.3: Calculate the output coefficient matrix W between the state variable x(k) and the output variable y(k) according to formula (1.8). out ∈j n ×4 : W out = inv(x × X T + β × I n ) × X × Y T Formula (1.8); In Equation (1.8), inv(·) represents the matrix inversion operation, the superscript T represents the vector transpose operation, β is an adjustable weighting coefficient, and I n ∈j n×n represents the identity matrix of size n; S1.4: Save the input coefficient matrix W in , the state transition coefficient matrix W, and the W in formula (1.8) out .

4. The optimal control method for a heat exchange station based on economic model predictive control according to claim 3, characterized in that: In formula (1.6), 500 ≤ n ≤ 1000, W in and W are both randomly generated within [-1, 1]; in formula (1.8), 10 -6 ≤ β ≤ 10 -2 .

5. The optimized control method for a heat exchange station based on economic model predictive control according to claim 3, characterized in that: In step S2, considering the heating demand of heat users and operation constraints, the method for designing an economic model predictive controller includes the following steps: S2.1: First, set the prediction step as P and the control step as M, and satisfy M ≤ P; Secondly, at the operation time l of the heat exchange station, M control sequences of the primary network water supply flow rate to be optimized and calculated are G1(l) = [G1(l + 1), G1(l + 2), …, G1(l + M)] Formula (2.1); M control sequences of the secondary network water supply flow rate to be optimized and calculated are G2(l) = [G2(l + 1), G2(l + 2), …, G2(l + M)] Formula (2.2); Then, based on the coefficient matrix W saved in step S1.4 in , W, and W out , calculate recursively the return water temperature T 1h (l + i|l), the return water pressure P 1h (l + i|l), the temperature difference △T2(l + i|l) between the supply and return water of the secondary network, and the pressure difference △P2(l + i|l) between the supply and return water of the secondary network, where i = 1, 2, …, P; Among them, T out (l + i) represents the predicted value of the outdoor temperature in the future time domain [l + 1, l + P], which is obtained from the weather forecast; where, ΔT2 is the temperature difference between the supply and return water of the secondary network, and ΔP2 is the pressure difference between the supply and return water of the secondary network; S2.2: Define the operation cost objective function J1 and the control increment objective function J2 of the heat exchange station respectively; Among them, in the objective function J1, c w is the specific heat capacity of water, J / (kg·℃); η1 and η2 respectively correspond to the price coefficient of purchased heat energy and the price coefficient of power consumption of the circulation pump, and when i > M, set G1(l + i) = G1(l + M) and G2(l + i) = G2(l + M); In the objective function J2, w1 and w2 are the weighting coefficients corresponding to the change amounts of the primary and secondary network water supply flow rates respectively; S2.3: Construct the following optimization problem with constraints; In formula (2.8), Q is the user's expected heat load in the future time domain [l+1, l+P], with the unit of KW; △P 1set is the lower limit of the pressure difference between the supply and return water of the primary heat supply network to ensure safe operation; △P 2set is the lower limit of the pressure difference between the supply and return water of the secondary heat supply network to ensure safe operation; S2.4: First, use the sequential quadratic programming method to solve the optimization problem described in step S2.3, and simultaneously obtain the optimal control sequences G1(l) and G2(l), and take the first items of the optimal control sequences G1(l) and G2(l) respectively, that is, G1(l + 1) and G2(l + 1); secondly, transmit them to the PLC control module of the heat exchange station through the background processor through the network for execution; then, transfer to the l + 1 moment, and repeat steps S2.1 to S2.4 to optimize and obtain the primary network water supply flow rate and the secondary network water supply flow rate, and execute the regulation of the heat exchange station.

6. A control system, characterized in that: Adopt the control method described in any one of claims 1 to 5; the control system includes a sensor group, a data storage module, an analysis and processing module, and a control module; The analysis and processing module includes a background processor; The control module includes the PLC control module of the heat exchange station; the PLC control module is connected to the background processor; The analysis and processing module is connected to the control module to transmit the obtained optimized heating parameters, and the control module is connected to the heating parameter control device of the heat exchange station to regulate the heating parameters of the heat exchange station according to the received optimized heating parameters; The sensor group is arranged on the water pipe and is used for monitoring and collecting the operation data of the heat exchange station; The data storage module, connected to the sensor group, is used for collecting and storing the real-time and historical operation data and meteorological data of the heat exchange station, and transmitting the operation data and meteorological data to the analysis and processing module; The analysis and processing module runs the control method according to the obtained data to calculate the optimized heating parameters.

7. The control system according to claim 6, wherein: The heating parameters include the primary network water supply flow rate and the secondary network water supply flow rate, and the heating parameter control device includes an electronic regulating valve, a frequency converter, and a circulating pump unit; The electronic regulating valve is arranged on the water pipe connecting the heat exchanger, the primary pipe network, and the secondary pipe network and is used for regulating the water flow rate.

8. The control system according to claim 6, characterized in that: The sensor group includes a pressure sensor, a flow sensor, and a temperature sensor.

9. The control system according to claim 8, wherein: The water pipes include a primary outlet pipe, a primary return pipe, a secondary outlet pipe, and a secondary return pipe; The temperature sensor, the pressure sensor, and the regulating valve are sequentially arranged on the primary outlet pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the secondary return pipe; the temperature sensor, the pressure sensor, and the valve are sequentially arranged on the secondary outlet pipe; the flow sensor, the pressure sensor, and the temperature sensor are sequentially arranged on the primary return pipe.

Images