An MPC intra-day scheduling optimization method for a combined cooling and power microgrid system

Abstract

This invention relates to an intraday MPC scheduling optimization method for combined cooling, heating, and power (CCHP) microgrid systems. The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intraday MPC scheduling optimization method for CCHP microgrid systems. The technical solution of this invention provides an intraday MPC scheduling optimization method for CCHP microgrid systems, comprising: S1, constructing transfer function models for each device in the microgrid system; S2, constructing a custom constraint set based on the energy constraints of each device; the energy constraints include capacity; S3, using the sum of the standard cost function and the custom cost function as the final optimization objective function; S4, setting appropriate time intervals, prediction time domains, and control time domains, and employing a sequential quadratic programming optimization algorithm to achieve MPC optimized scheduling calculations. This invention is applicable to the field of energy dispatching technology.

Description

Technical Field

[0001] This invention relates to the field of energy dispatching technology, and more specifically, to an MPC intraday dispatching optimization method for combined cooling, heating and power microgrid systems. Background Technology

[0002] Combined cooling, heating, and power (CCHP) microgrid systems have the advantages of high primary energy utilization and low environmental pollution impact, making them a hot topic of interest in recent years. A CCHP microgrid system can simultaneously generate cooling, heating, and electricity, achieving a primary energy utilization rate of 75%-80%, while consuming only 3 / 4 of the energy required for traditional separate heating and power systems. A typical CCHP microgrid system includes (1) generator sets, such as gas turbines, internal combustion engines, and fuel cells; (2) heating equipment, such as waste heat boilers and gas boilers; and (3) refrigeration equipment, such as lithium bromide absorption chillers and electric chillers.

[0003] Due to the influence of outdoor factors such as climate change and indoor factors such as people and equipment, the load exhibits significant uncertainty and volatility. From the time each cooling and heating equipment adjusts its scheduling plan according to demand load to the point where the cooling and heating received by the end user stabilizes, this process typically involves a relatively long transition time, specifically related to the dynamic response characteristics of the entire system. The dynamic response time of an integrated energy cooling and heating system is mainly affected by the system's heat capacity, the length of the transmission pipelines, the heat capacity of the radiators, and the type of building envelope materials. Different integrated energy systems have different dynamic response times; generally, the dynamic response time may be 5-8 hours, and for more complex systems, it may be even longer. If scheduling optimization is performed only based on the system's static characteristics, without considering the dynamic process of energy transfer, it will lead to a mismatch between supply and demand, failing to effectively match daily fluctuations in demand and resulting in decreased user satisfaction. Therefore, it is necessary to consider the dynamic process of energy transfer, establish a dynamic model of the cooling and heating equipment input to the user end, and perform scheduling optimization calculations based on this model. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and to propose an MPC intraday scheduling optimization method for combined cooling, heating and power microgrid systems.

[0005] Firstly, a method for optimizing intraday scheduling of combined cooling, heating, and power (CCHP) microgrid systems is provided, including:

[0006] S1. Construct the transfer function model of each device in the microgrid system;

[0007] S2. Construct a custom constraint set based on the energy constraints of each device; the energy constraints include capacity.

[0008] S3. The sum of the MPC standard cost function and the custom cost function is used as the final optimization objective function;

[0009] S4. Set appropriate time intervals, prediction time domains, and control time domains, and use a sequential quadratic programming optimization algorithm to achieve MPC optimized scheduling calculation.

[0010] Preferably, in S1, the microgrid system includes: a power grid, a prime mover, a waste heat boiler, an absorption chiller, a gas boiler, and an electric chiller; the electrical energy generated by the prime mover is transmitted through wires to an external power supply network that supplies power to users, and the power grid supplements the power supply when the prime mover's power supply is insufficient; the heat energy generated by the prime mover is transmitted through pipelines to the waste heat boiler and the absorption chiller respectively; the heat generated by the waste heat boiler is transmitted to a heating network that supplies heat to users; the cold energy generated by the absorption chiller is transmitted to a cooling network that supplies cooling to users; the heat energy generated by the gas boiler is transmitted to the heating network to supplement the heating supply when the waste heat boiler's heating supply is insufficient; an electric chiller is connected to the power supply network, and the cold energy generated by the electric chiller is transmitted through pipelines to the cooling network to supplement the cooling supply when the absorption chiller's cooling supply is insufficient.

[0011] Preferably, in S1, the heating and cooling pipe networks of the microgrid system are both equipped with primary and secondary networks, and energy exchange between the primary and secondary networks is achieved through cold and hot plate exchange stations.

[0012] As a preferred embodiment, in S1, excitation data is designed based on the characteristics of the equipment, input to the input terminals of each equipment, and output data is obtained. Then, based on the input and output data, the transfer function model of each equipment in the microgrid system is constructed using the system identification method.

[0013] As a preferred embodiment, in S1, excitation data is designed based on the characteristics of the equipment, input to the input terminals of each equipment, and output data is obtained. Then, based on the input and output data, the transfer function model of each equipment in the microgrid system is constructed using the system identification method.

[0014] Preferably, in S1, the equipment characteristic refers to the step response time of each cooling and heating device.

[0015] Preferably, in S3, the MPC standard objective function is determined by the following formula:

[0016] ;

[0017] in, For the standard objective function, To track deviations from the reference output, To track the deviation of the control variables, To suppress the increment of the control variable, This is the penalty function term resulting from the violation of constraints. Without considering control variable tracking and control variable increment suppression, and assuming no other soft constraints, , , All are 0. Therefore, the standard objective function of MPC can be expressed as:

[0018] ;

[0019] Where k is the current control time interval; p is the prediction time domain; To output the number of variables; Let j be the value of the output predicted at the i-th prediction time step. This is the j-th output reference value at the i-th prediction time step; The scaling factor for the j-th output; Let j be the output weight for the i-th prediction time step; The QP solution to the optimization problem is expressed as:

[0020] .

[0021] Preferably, in S3, the custom cost function is expressed as:

[0022] ;

[0023] Among them, C gas Natural gas cost, in yuan; C ele C represents the cost of purchasing electricity from the power grid, expressed in yuan. om Operation and maintenance costs are expressed in yuan; Ts is the time interval, expressed in seconds; R gas This refers to the price of natural gas, expressed in yuan / kg; R ele This refers to the time-of-use electricity price, expressed in yuan / kWh; R gt R gb R ec R ac R b The prices listed are for the operation and maintenance of gas turbines, gas boilers, electric chillers, absorption chillers, and waste heat boilers, in yuan / kWh.

[0024] The final optimization objective function is expressed as:

[0025] .

[0026] Secondly, a daily scheduling optimization device for a combined cooling, heating, and power (CCHP) microgrid system is provided, used to execute the daily scheduling optimization method for a CCHP microgrid system as described in any of the first aspects, including:

[0027] The first construction module is used to build the transfer function models of each device in the microgrid system;

[0028] The second construction module is used to construct a custom constraint set based on the energy constraints of each device; the energy constraints include capacity.

[0029] The function definition module is used to take the sum of the MPC standard objective function and the custom cost function as the final optimization objective function;

[0030] The calculation module is used to set appropriate time intervals, prediction time domains, and control time domains, and uses a sequential quadratic programming optimization algorithm to achieve MPC optimized scheduling calculation.

[0031] Thirdly, a computer storage medium is provided, wherein a computer program is stored in the computer storage medium; when the computer program is run on a computer, the computer executes the MPC intraday scheduling optimization method for a combined cooling, heating and power microgrid system as described in any of the first aspects.

[0032] The beneficial effects of this invention are as follows: This invention designs an MPC intraday scheduling optimization method for combined cooling, heating, and power (CCHP) microgrid systems. Through system identification and modeling methods, a dynamic model of the transfer function of each cooling and heating device within the system is established. Based on this, the MPC rolling optimization method is used to optimize the intraday scheduling of each device, and the scheduling plan is determined based on the optimization results. Compared with previous scheduling optimization methods based on the static characteristics of equipment, this invention considers the dynamic energy transfer process of the system, optimizes and adjusts the scheduling plan based on static characteristics, and can better match the system's scheduling output with the changing cooling and heating demands of users, improving user satisfaction and minimizing the overall operating cost of the system. Attached Figure Description

[0033] Figure 1 A schematic diagram of a combined cooling, heating and power microgrid system;

[0034] Figure 2 A schematic diagram of the excitation data for modeling the heating section;

[0035] Figure 3 This is a schematic diagram comparing the output of the heating model established with the actual heating output of the APROS system when test data is used as input.

[0036] Figure 4A schematic diagram of the Simulink graphical program for the MPC scheduling optimization algorithm;

[0037] Figure 5 A comparison of the cooling capacity output of the APROS simulation system under the MPC-optimized scheduling strategy and the static characteristic-based scheduling strategy;

[0038] Figure 6 A comparison of the heat output results of the APROS simulation system under the MPC-optimized scheduling strategy and the static characteristic-based scheduling strategy;

[0039] Figure 7 A schematic diagram comparing the scheduling plans of electric chillers under the MPC-optimized scheduling algorithm and the scheduling algorithm based on static characteristics;

[0040] Figure 8 A schematic diagram comparing the scheduling plans of the absorption chiller under the MPC-optimized scheduling algorithm and the static characteristic-based scheduling algorithm;

[0041] Figure 9 A schematic diagram comparing the waste heat boiler scheduling plans under the MPC-optimized scheduling algorithm and the static characteristic-based scheduling algorithm;

[0042] Figure 10 A schematic diagram comparing the gas boiler scheduling plans under the MPC-optimized scheduling algorithm and the static characteristic-based scheduling algorithm. Detailed Implementation

[0043] The present invention will be further described below with reference to embodiments. The description of the embodiments below is only for the purpose of helping to understand the present invention. It should be noted that those skilled in the art can make several modifications to the present invention without departing from the principle of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

[0044] Example 1:

[0045] like Figure 1 As shown, the microgrid system provided in this application includes: a power grid, a prime mover, a waste heat boiler, an absorption chiller, a gas boiler, and an electric chiller. The prime mover receives natural gas flow (gas1); the gas boiler receives natural gas flow (gas2); P0 is the power output of the gas turbine; the gas turbine outputs heat (Qex); the absorption chiller receives waste heat (Qex1); the waste heat boiler receives waste heat (Qex2); P2 is supplementary power from the power grid; P1 is the input power of the electric chiller; the electric chiller outputs cooling capacity (cold1); the absorption chiller outputs cooling capacity (cold2); the waste heat boiler outputs heat (heat1); and the gas boiler outputs heat (heat2).

[0046] The prime mover is a gas turbine, which generates heat or electricity by consuming natural gas. The electricity generated by the gas turbine is transmitted via wires to the external power grid that supplies electricity to users, providing electrical load to the system. The system can also meet its electricity demand by purchasing electricity from the grid. The heat energy generated by the gas turbine is piped to a waste heat boiler and an absorption chiller.

[0047] Waste heat boilers generate heat from hot flue gas to provide a heat load for the system, and the generated heat is then transported to heating pipelines that supply heat to users. Absorption chillers generate cooling capacity from hot flue gas to provide a cooling load, and the generated cooling capacity is then transported to cooling pipelines that supply cooling to users.

[0048] Gas-fired boilers generate heat by consuming natural gas, compensating for insufficient heating from waste heat boilers. Electric chillers are connected to the power grid, and the cooling generated by these chillers is supplied to the cooling pipelines, supplementing the cooling capacity of the absorption chillers. Furthermore, both the heating and cooling pipelines of the microgrid system are divided into primary and secondary networks, with energy exchange between them achieved through cold and heat plate exchange stations.

[0049] Due to the lack of actual cooling and heating systems, this invention focuses on scheduling optimization research based on a cooling and heating system built on APROS simulation software. The simulation system is designed according to... Figure 1 The system can be constructed in this manner. For cooling and heating systems already in operation in actual engineering projects, the scheduling and optimization methods provided by this invention can also be used.

[0050] Example 2:

[0051] Considering that the dynamic response process of the power supply section is shorter compared to the cooling and heating sections, the MPC intraday scheduling optimization method provided by this invention focuses on optimizing the scheduling of the cooling and heating sections, including:

[0052] S1. Construct the transfer function model of each device in the microgrid system.

[0053] In S1, excitation data is designed based on the equipment characteristics and input to the input terminals of each device to obtain output data. Then, based on the input and output data, the transfer function model of each device in the microgrid system is constructed using a system identification method. The equipment characteristics mainly refer to the step response time of each cooling and heating device.

[0054] Taking the heating section as an example, this paper illustrates the method for establishing a transfer function model for the equipment. Through step response testing, the dynamic response process of the gas-fired boiler and waste heat boiler in the constructed simulation system is approximately 4 hours. That is, from the change of input on the heat source side to the end user obtaining stable heat, the entire process takes approximately 4 hours. The excitation data uses a GBN signal with an average transition time of one-third of the dynamic response time, i.e., a GBN signal with an average transition time of 4800s is used as the excitation test signal for the gas-fired boiler and waste heat boiler. The data sampling time is 1s, and the excitation duration can be designed to be approximately 6-18 times the entire dynamic response time.

[0055] The final test signal input and the obtained output are shown in the figure. Figure 2 As shown, from top to bottom, the test inputs are for the waste heat boiler, the test inputs for the gas boiler, and the total heat output of the actual system.

[0056] Then, system identification is performed based on the input and output. Modeling can be done using the MATLAB System Identification Toolbox, and the final transfer function model of the heating section is shown in the following equation:

[0057] ;

[0058] HEAT represents the total system heat supply.

[0059] The accuracy of the model is determined by Figure 3 It is evident that this can accurately reflect the dynamic characteristics of the actual heating system.

[0060] The refrigeration section was modeled using the same method.

[0061] S2. Construct a custom constraint set based on the energy constraints of each device; the energy constraints include capacity, and all custom constraints are in the form of hard constraints.

[0062] In this embodiment, the capacity constraints of each device are shown in the following formula:

[0063] 0≤cold1≤1364, 0≤cold2≤924, 0≤heat1≤352, 0≤heat2≤242

[0064] P1≥0, Qex1≥0, Qex2≥0, gas2≥0

[0065] In addition, the law of conservation of energy must be satisfied, as shown in the following equation:

[0066] Qex1 + Qex2 = Qex

[0067] S3. The sum of the standard MPC objective function and the user-defined cost function is used as the final optimization objective function. The standard objective function is used to balance the matching degree between system scheduling output and user load demand, while the user-defined cost function is used to measure the economy of the scheduling plan.

[0068] In S3, the standard objective function for MPC is determined by the following formula:

[0069] ;

[0070] in, For the standard objective function, To track deviations from the reference output, To track the deviation of the control variables, To suppress the increment of the control variable, This is the penalty function term resulting from the violation of constraints. Without considering control variable tracking and control variable increment suppression, and assuming no other soft constraints, , , All are 0. Therefore, the standard objective function of MPC can be expressed as:

[0071] ;

[0072] Where k is the current control time interval; p is the prediction time domain; To output the number of variables; Let j be the value of the output predicted at the i-th prediction time step. This is the j-th output reference value at the i-th prediction time step; The scaling factor for the j-th output; Let j be the output weight for the i-th prediction time step; The QP solution to the optimization problem is expressed as:

[0073] .

[0074] The custom cost function is expressed as:

[0075] ;

[0076] Among them, C gas Natural gas cost, in yuan; C ele C represents the cost of purchasing electricity from the power grid, expressed in yuan. om Operation and maintenance costs are expressed in yuan; Ts is the time interval, expressed in seconds; R gas This refers to the price of natural gas, expressed in yuan / kg; R ele This refers to the time-of-use electricity price, expressed in yuan / kWh; R gt Rgb R ec R ac R b The following are the operation and maintenance prices for gas turbine, gas boiler, electric chiller, absorption chiller, and waste heat boiler, respectively, in yuan / kWh; gas1 is the natural gas flow rate input to the prime mover; gas2 is the natural gas flow rate input to the gas boiler; P0 is the power output of the gas turbine; P2 is the supplementary power supply from the power grid; P1 is the input power of the electric chiller; cold1 is the cooling capacity output of the electric chiller; cold2 is the cooling capacity output of the absorption chiller; heat1 is the heat output of the waste heat boiler.

[0077] The final optimization objective function is expressed as:

[0078] .

[0079] S4. Set appropriate time intervals (sampling time), prediction time domain and control time domain, and use a sequential quadratic programming optimization algorithm to achieve MPC optimized scheduling calculation.

[0080] This invention utilizes the MATLAB Nonlinear MPC Toolbox to implement all optimization algorithms and programs. An MPC prediction model is constructed using the established system transfer function model. Custom MPC constraints are built using the constraint set established in S2. A custom cost function is constructed using the objective function form described in S3. These are then combined with the standard objective function provided by the toolbox to construct the final optimization objective function. The optimization calculation is performed using the default sequential quadratic programming algorithm of the MATLAB Nonlinear MPC Toolbox (the toolbox uses the built-in MATLAB function `fmincon` for optimization). In this embodiment, through experiments, the final determined MPC sampling time is 15 minutes, and both the prediction time domain and control time domain are 8 minutes (i.e., the prediction duration and control duration are both 8 * 15 minutes).

[0081] Finally, a graphics computing program was built in the Simulink interface, see... Figure 4 As shown. After running the calculation program, scheduling optimization instructions are obtained. These instructions are then transmitted to the APROS simulation system to obtain the final system cooling and heating outputs. The results are compared with previous optimization scheduling results based on the system's static characteristics. The final results are shown below. Figure 5 and Figure 6As shown in Figures 7, 8, 9, and 10. The results indicate that during periods of significant change in user load demand, the scheduling method using the system dynamic model is significantly better than the scheduling method using the static model in matching the cold and hot load scheduling output with the user's predicted load demand. This is because the dynamic model can reflect the dynamic characteristics of the system's heat transfer. Compared to scheduling strategies utilizing static characteristics, this scheme can additionally consider the impact of dynamic processes, allowing for further optimization and improvement of the original scheduling scheme. The results of the improved scheduling strategy are shown in Figures 7, 8, 9, and 10.

[0082] The method described above for implementing integrated energy system scheduling optimization using the MATLAB nonlinear MPC toolbox is only one means of implementing the MPC optimization scheduling algorithm, and is not intended to limit the present invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all technical solutions obtained by equivalent substitution or equivalent transformation fall within the protection scope of the present invention.

Claims

1. A method for intraday scheduling optimization of MPC (Multi-Processing Control) in a combined cooling, heating, and power (CCHP) microgrid system, characterized in that, include: S1. Construct the transfer function model of each device in the microgrid system; S2. Construct a custom constraint set based on the energy constraints of each device; the energy constraints include capacity. S3. The sum of the MPC standard objective function and the user-defined cost function is used as the final optimization objective function; S4. Set appropriate time intervals, prediction time domains, and control time domains, and use a sequential quadratic programming optimization algorithm to achieve MPC optimized scheduling calculation; In S3, the MPC standard objective function is determined by the following formula: ； in, For the standard objective function, To track deviations from the reference output, To track the deviation of the control variables, To suppress the increment of the control variable, This is the penalty function term resulting from the violation of constraints; without considering control variable tracking and control variable increment suppression, and without other soft constraints, , , If all values ​​are 0, then the standard objective function of MPC can be expressed as: ； Where k is the current control time interval; p is the prediction time domain; To output the number of variables; Let j be the value of the output predicted at the i-th prediction time step. This is the j-th output reference value at the i-th prediction time step; The scaling factor for the j-th output; Let j be the output weight for the i-th prediction time step; The QP solution to the optimization problem is expressed as: 。 2. The MPC intraday scheduling optimization method for combined cooling, heating and power microgrid systems according to claim 1, characterized in that, In S1, the microgrid system includes: a power grid, a prime mover, a waste heat boiler, an absorption chiller, a gas boiler, and an electric chiller. The electrical energy generated by the prime mover is transmitted via wires to an external power supply network that supplies electricity to users; when the prime mover's power supply is insufficient, it is supplemented by the power grid. The heat energy generated by the prime mover is transmitted via pipelines to the waste heat boiler and the absorption chiller, respectively. The heat generated by the waste heat boiler is transmitted to a heating network that supplies heat to users. The cold energy generated by the absorption chiller is transmitted to a cooling network that supplies cooling to users. The heat energy generated by the gas boiler is transmitted to the heating network to supplement heating when the waste heat boiler's heating supply is insufficient. An electric chiller is connected to the power supply network, and the cold energy generated by the electric chiller is transmitted via pipelines to the cooling network to supplement cooling when the absorption chiller's cooling supply is insufficient.

3. The MPC intraday scheduling optimization method for combined cooling, heating, and power microgrid systems according to claim 2, characterized in that, In S1, the heating and cooling pipe networks of the microgrid system are both equipped with primary and secondary networks, and energy exchange between the primary and secondary networks is achieved through cold and hot plate exchange stations.

4. The MPC intraday scheduling optimization method for combined cooling, heating and power microgrid systems according to claim 1, characterized in that, In S1, excitation data is designed based on the characteristics of the equipment, input to the input terminals of each equipment, and output data is obtained. Then, based on the input and output data, the transfer function model of each equipment in the microgrid system is constructed using the system identification method.

5. The MPC intraday scheduling optimization method for a combined cooling, heating, and power microgrid system according to claim 4, characterized in that, In S1, the equipment characteristics refer to the step response time of each cooling and heating device.

6. The MPC intraday scheduling optimization method for a combined cooling, heating, and power microgrid system according to claim 1, characterized in that, In S3, the custom cost function is expressed as: ； Among them, C gas Natural gas cost, in yuan; C ele C represents the cost of purchasing electricity from the power grid, expressed in yuan. om Operation and maintenance costs are expressed in yuan; Ts is the time interval, expressed in seconds; R gas This refers to the price of natural gas, expressed in yuan / kg; R ele This refers to the time-of-use electricity price, expressed in yuan / kWh; R gt R gb R ec R ac R b The following are the operation and maintenance prices for the prime mover, gas boiler, electric chiller, absorption chiller, and waste heat boiler, respectively, in yuan / kWh; gas1 is the natural gas flow rate input to the prime mover; gas2 is the natural gas flow rate input to the gas boiler; P0 is the power output of the gas turbine; P2 is the supplementary power supply from the power grid; P1 is the input power of the electric chiller; cold1 is the cooling capacity output of the electric chiller; cold2 is the cooling capacity output of the absorption chiller; heat1 is the heat output of the waste heat boiler; heat2 is the heat output of the gas boiler. The final optimization objective function is expressed as: 。 7. An MPC intraday scheduling optimization device for a combined cooling, heating and power microgrid system, characterized in that, The method for executing the MPC intraday scheduling optimization method for a combined cooling, heating and power microgrid system as described in any one of claims 1 to 6 includes: The first construction module is used to build the transfer function models of each device in the microgrid system; The second construction module is used to construct a custom constraint set based on the energy constraints of each device; the energy constraints include capacity. The function definition module is used to take the sum of the MPC standard objective function and the custom cost function as the final optimization objective function; The calculation module is used to set appropriate time intervals, prediction time domain, and control time domain, and adopts a sequential quadratic programming optimization algorithm to realize MPC optimized scheduling calculation; The standard objective function of MPC is determined by the following formula: ； in, For the standard objective function, To track deviations from the reference output, To track the deviation of the control variables, To suppress the increment of the control variable, This is the penalty function term resulting from the violation of constraints; without considering control variable tracking and control variable increment suppression, and without other soft constraints, , , If all values ​​are 0, then the standard objective function of MPC can be expressed as: ； Where k is the current control time interval; p is the prediction time domain; To output the number of variables; Let j be the value of the output predicted at the i-th prediction time step. This is the j-th output reference value at the i-th prediction time step; The scaling factor for the j-th output; Let j be the output weight for the i-th prediction time step; The QP solution to the optimization problem is expressed as: 。 8. A computer storage medium, characterized in that, The computer storage medium stores a computer program; when the computer program is run on the computer, it causes the computer to execute the MPC intraday scheduling optimization method for a combined cooling, heating and power microgrid system as described in any one of claims 1 to 6.

Images