Fuel gas and thermal power generation combined power supply control method, device, equipment and medium

Abstract

The invention relates to the technical field of electric power systems, in particular to a gas and thermal power generation combined power supply control method, device and equipment and a medium, and the method comprises the steps: obtaining historical power consumption data, and carrying out the power demand prediction through employing a long short-term memory (LSTM) neural network model based on the historical power consumption data, and obtaining a prediction result; according to a prediction result, optimizing output distribution of gas power generation and thermal power generation by using a mixed integer linear programming MILP method to obtain a scheduling model; setting different power generation scenes, solving optimal scheduling schemes of the scheduling model in a long period and multiple scenes, evaluating economic benefits of each energy output in different optimal scheduling schemes, and selecting an optimal scheduling scheme; the power load change is monitored in real time, and the output of gas power generation and thermal power generation is adjusted in real time by taking the optimal scheduling scheme as a reference and adopting a PID control algorithm. According to the invention, through demand prediction, resource scheduling, real-time control and economic analysis, the operation economy and stability of the power system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a control method, device, equipment and medium for combined power supply of gas and thermal power generation. Background Art

[0002] In modern power systems, gas-fired power generation and thermal power generation are the two main power generation methods. Gas-fired power generation uses natural gas, biogas, etc. to generate electricity. It has fast power generation speed, high efficiency, and low pollution. It is a clean energy power generation method. Thermal power generation uses fossil fuels such as coal and oil to generate electricity. It has low power generation cost and is suitable for large power stations and fixed large loads. Although traditional thermal power generation has a relatively stable power generation, it has high carbon dioxide emissions. It needs to be combined with clean energy power generation methods such as gas-fired power generation to reduce the impact on the environment. Although gas-fired power generation has the advantages of fast start-up and flexible regulation, its gas supply is uncertain and the price is high. Therefore, how to reduce the impact of power generation methods on the environment and improve their flexibility under the premise of ensuring stable power generation demand and economy; how to effectively combine the advantages of gas-fired power generation and thermal power generation to improve the operating efficiency and economic benefits of the power system have become hot issues in current research.

[0003] At present, the existing technology has predicted the cost of the comprehensive power generation system composed of gas-fired power generation and thermal power generation, but has not analyzed its economic feasibility in combination with the power generation cost and electricity cost; there are also related technologies that have studied the use of thermal power generation exhaust gas as a gas source for gas-fired power generation, but have not yet specifically introduced the impact of the output of the two power generation methods on the power generation demand of the power grid. In summary, although the existing technology attaches importance to the combination of gas-fired power generation and thermal power generation, it has not controlled the output of the two power generation strategies from the perspective of demand forecasting, optimal scheduling, and economic analysis, and there are problems such as loose combination of power generation methods and insufficient economic considerations.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0005] The present invention provides a control method, device, equipment and medium for combined power supply of gas and thermal power generation, thereby effectively solving the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: a control method for combined power supply of gas and thermal power generation, comprising the following steps:

[0007] Obtain historical electricity consumption data, and use a long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain a prediction result;

[0008] According to the prediction results, the output distribution of gas-fired power generation and thermal power generation is optimized using the mixed integer linear programming MILP method to obtain a scheduling model;

[0009] Set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme;

[0010] The power load changes are monitored in real time, and the output of gas-fired power generation and thermal power generation is adjusted in real time using a PID control algorithm with reference to the optimal dispatching plan.

[0011] Furthermore, the power demand forecasting using the long short-term memory (LSTM) neural network model includes the following steps:

[0012] Normalizing the power consumption and total power consumption in the historical power consumption data;

[0013] The normalized power consumption and total power consumption data are used as input of the LSTM neural network model;

[0014] Based on the LSTM unit structure and network connection structure, complete the setting of multiple different LSTM neural network hyperparameters;

[0015] Run the LSTM neural network model to predict power consumption and total power consumption.

[0016] Furthermore, the normalizing of the power consumption and the total power consumption in the historical power consumption data includes:

[0017]

[0018] In the formula, P and W are the power consumption and total power consumption at the current moment, respectively. min and P max are the minimum and maximum values ​​of power consumption, W max and W min are the minimum and maximum values ​​of total power consumption, P user and W user are the normalized results respectively.

[0019] Furthermore, the LSTM neural network model is run to predict the power consumption and total power consumption, and the calculation formula is as follows:

[0020] F=LSTM(epoch,batch-size,dropout);

[0021] P′ user =F(P user );

[0022] W′ user =F(W user );

[0023] Where F represents the LSTM neural network prediction model, epoch, batch-size and dropout are all network coefficients, which are determined according to the network parameters; P′ user and W′ user are the power consumption and total power consumption calculated by the LSTM neural network prediction model, P user and W user are the normalized results respectively.

[0024] Furthermore, the mixed integer linear programming MILP method is used to optimize the output distribution of gas power generation and thermal power generation to obtain a scheduling model, including:

[0025] Taking the minimum operating cost as the objective function, a dispatching model is constructed. The dispatching model includes two subsystems, the power system and the natural gas system. The gas units in the natural gas system are coupled to participate in the power supply peak regulation, and the operating cost is minimized by optimizing the output distribution of gas-fired power generation and thermal power generation;

[0026] Establish multi-dimensional constraints that take into account power and natural gas flow balance, equipment operation restrictions, and grid security;

[0027] Convex optimization is performed on nonlinear terms in the scheduling model to convert it into a mixed integer linear programming problem;

[0028] The solver is called to solve the mixed integer linear programming problem, and under the premise of satisfying the constraints, the optimal output allocation scheme of gas power generation and thermal power generation and the corresponding start and stop plan of the generator set are determined.

[0029] Furthermore, the objective function C ope include:

[0030]

[0031] In the formula, C gas , C store , C gc , C g , C w , C essThey are the gas source gas output cost, gas tank operation cost, thermal power unit operation cost, wind turbine unit operation cost, gas turbine operation cost and power shortage and power restriction penalty cost respectively; T is the scheduling optimization cycle, T = 24h; set is the total number of model scheduling scenarios; i, st, gc, g, w are the nodes where the gas source, gas tank, coal-fired unit, gas unit and wind turbine are located respectively; gen is the node where each generator set in the system is located; Ω s ,Ω gas ,Ω g ,Ω c ,Ω store and Ω gen They are typical day scenes, natural gas source nodes, gas turbine units, coal-fired generator units, gas storage equipment, and the collection of all generator unit equipment; G s,i,source is the total gas consumption of node i on a typical day; c i,gas is the contract gas purchase price at node i; c gc 、c w With c g They are the operation and maintenance costs per unit output of thermal power generating units, fans and gas turbines; c st,in 、c st,out They are the gas inlet and outlet costs per unit flow of the gas storage tank; Generate penalty costs for power shortage and power abandonment in the system; as well as They are the outputs of thermal power units, fans and gas turbines at time s and t on a typical day respectively; They are the gas inlet and outlet volumes of the gas storage tank respectively; It is the sum of the output of each generator set in the system; is the active power consumed by the load at time s, t on a typical day.

[0032] Furthermore, the performing convex optimization on the nonlinear terms in the scheduling model includes:

[0033] Linearize the flow calculation equations in the natural gas system:

[0034]

[0035]

[0036] In the formula, is the tidal flow rate from m to n in the gas distribution network transmission pipeline at time t; and are the gas pressures of natural gas nodes m and n at time s and t on a typical day; K mn Weymouth characteristic parameters of natural gas pipeline.

[0037] Further, the calling solver to solve the mixed integer linear programming problem includes:

[0038] The prediction result is used as the active load to solve for the specific value of .

[0039] Furthermore, the evaluation of the economic benefits of each energy output under different optimization scheduling schemes includes:

[0040] The economic benefits of different scheduling strategies include direct costs and indirect benefits. The direct costs include energy consumption, equipment wear and tear, and operation and maintenance costs; the indirect benefits include energy utilization optimization and carbon emissions.

[0041] Combining the internal rate of return method, net present value analysis, sensitivity analysis and regression period analysis methods, the economic feasibility of different scheduling strategies is evaluated in multiple dimensions;

[0042] Game theory is introduced, and power generation companies randomly initialize their own bidding strategies within the upper and lower limits of the bidding strategies. The power generation and purchase costs, power generation profits, and energy utilization trend curves under various scenarios are obtained through multiple iterations.

[0043] The economic benefits and future development trends of each energy output under different scheduling strategies are obtained.

[0044] Furthermore, the PID control algorithm is used to adjust the output of gas-fired power generation and thermal power generation in real time, including:

[0045]

[0046] Where, e(t) is the error signal of a single power generation system state, and the error signal is the error between the real-time output signal and the reference output signal; Δe(t) is the error signal change rate; K P , K I , K D are proportional, integral and differential coefficients; t is time.

[0047] The present invention also includes a control device for combined power supply of gas and thermal power generation, using the above method, the device includes:

[0048] A prediction unit, used to obtain historical electricity consumption data, and use a long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain a prediction result;

[0049] A modeling unit, configured to optimize the output distribution of gas-fired power generation and thermal power generation by using a mixed integer linear programming (MILP) method according to the prediction results to obtain a scheduling model;

[0050] An evaluation unit is used to set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme;

[0051] The control unit is used to monitor the change of power load in real time, and to adjust the output of gas power generation and thermal power generation in real time by using PID control algorithm with reference to the optimal dispatching scheme.

[0052] The present invention also includes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described above is implemented.

[0053] The present invention also includes a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0054] The beneficial effects of the present invention are: improving the economy and stability of power system operation through demand forecasting, resource scheduling, real-time control and economic analysis. The strategy uses big data and machine learning algorithms to forecast demand, adopts optimization methods for resource scheduling, and ensures the stable operation of the power system through real-time monitoring and adjustment. Through economic analysis, the optimal scheduling scheme is selected to achieve the lowest operating cost and the maximum benefit. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 is a flow chart of the method in Example 1;

[0057] Figure 2 It is a structural schematic diagram of the device in Example 1;

[0058] Figure 3 This is a flowchart of demand forecasting in Example 2;

[0059] Figure 4 This is a flowchart of resource scheduling in Example 2;

[0060] Figure 5 This is a flowchart of real-time control in Example 2;

[0061] Figure 6 is a flowchart of the economic analysis in Example 2;

[0062] Figure 7 It is a schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0064] Embodiment 1:

[0065] like Figure 1 As shown: A control method for combined power supply of gas and thermal power generation, comprising the following steps:

[0066] Obtain historical electricity consumption data, and use the long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain prediction results;

[0067] According to the prediction results, the mixed integer linear programming (MILP) method is used to optimize the output distribution of gas-fired power generation and thermal power generation, and the scheduling model is obtained;

[0068] Set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme;

[0069] Monitor the changes in power load in real time, use the optimal dispatch plan as a reference, and use PID control algorithm to adjust the output of gas-fired power generation and thermal power generation in real time.

[0070] Improve the economy and stability of power system operation through demand forecasting, resource scheduling, real-time control and economic analysis. This strategy uses big data and machine learning algorithms for demand forecasting, uses optimization methods for resource scheduling, and ensures the stable operation of the power system through real-time monitoring and adjustment. Through economic analysis, the optimal scheduling scheme is selected to achieve the lowest operating cost and maximum benefit.

[0071] In this embodiment, a long short-term memory (LSTM) neural network model is used to predict power demand, including the following steps:

[0072] Normalize the power consumption and total power consumption in the historical power consumption data;

[0073] The normalized power consumption and total power consumption data are used as inputs to the LSTM neural network model;

[0074] Based on the LSTM unit structure and network connection structure, complete the setting of multiple different LSTM neural network hyperparameters;

[0075] Run the LSTM neural network model to predict power consumption and total power consumption.

[0076] Normalize the power consumption and total power consumption in the historical power consumption data, including:

[0077]

[0078] In the formula, P and W are the power consumption and total power consumption at the current moment, respectively. min and P max are the minimum and maximum values ​​of power consumption, W max and W min are the minimum and maximum values ​​of total power consumption, P user and W user are the normalized results respectively.

[0079] Among them, the LSTM neural network model is run to predict the power consumption and total power consumption, and its calculation formula is as follows:

[0080] F=LSTM(epoch,batch-size,dropout);

[0081] P′ user =F(P user );

[0082] W′ user =F(W user );

[0083] Where F represents the LSTM neural network prediction model, epoch, batch-size and dropout are all network coefficients, which are determined according to the network parameters; P′ user and W′ user are the power consumption and total power consumption calculated by the LSTM neural network prediction model, P user and W user are the normalized results respectively.

[0084] As a preferred embodiment of the above, a mixed integer linear programming MILP method is used to optimize the output distribution of gas power generation and thermal power generation to obtain a scheduling model, including:

[0085] Taking the minimum operating cost as the objective function, a dispatching model is constructed. The dispatching model includes two subsystems: the power system and the natural gas system. The gas units in the natural gas system are coupled to participate in the peak load regulation of power supply, and the operating cost is minimized by optimizing the output distribution of gas-fired power generation and thermal power generation.

[0086] Establish multi-dimensional constraints that take into account power and natural gas flow balance, equipment operation restrictions, and grid security;

[0087] The nonlinear terms in the scheduling model are convexly optimized and converted into a mixed integer linear programming problem;

[0088] The solver is called to solve the mixed integer linear programming problem. Under the premise of meeting the constraints, the optimal output allocation plan for gas-fired power generation and thermal power generation, as well as the corresponding start and stop plan of the generator sets are determined.

[0089] Objective function C ope include:

[0090]

[0091] In the formula, C gas , C store , C gc , C g , C w , C ess They are the gas source gas output cost, gas tank operation cost, thermal power unit operation cost, wind turbine unit operation cost, gas turbine operation cost and power shortage and power restriction penalty cost respectively; T is the scheduling optimization cycle, T = 24h; set is the total number of model scheduling scenarios; i, st, gc, g, w are the nodes where the gas source, gas tank, coal-fired unit, gas unit and wind turbine are located respectively; gen is the node where each generator set in the system is located; Ω s ,Ω gas ,Ω g ,Ω c ,Ω store and Ω gen They are typical day scenes, natural gas source nodes, gas turbine units, coal-fired generator units, gas storage equipment, and the collection of all generator unit equipment; G s,i,source is the total gas consumption of node i on a typical day; c i,gas is the contract gas purchase price at node i; c gc 、c w With c g They are the operation and maintenance costs per unit output of thermal power generating units, fans and gas turbines; c st,in 、c st,out They are the gas inlet and outlet costs per unit flow of the gas storage tank; Generate penalty costs for power shortage and power abandonment in the system; as well as They are the outputs of thermal power units, fans and gas turbines at time s and t on a typical day respectively; They are the gas inlet and outlet volumes of the gas storage tank respectively; It is the sum of the output of each generator set in the system; is the active power consumed by the load at time s, t on a typical day.

[0092] In this embodiment, the nonlinear terms in the scheduling model are convexly optimized, including:

[0093] Linearize the flow calculation equations in the natural gas system:

[0094]

[0095]

[0096] In the formula, is the tidal flow rate from m to n in the gas distribution network transmission pipeline at time t; and are the gas pressures of natural gas nodes m and n at time s and t on a typical day; K mn Weymouth characteristic parameters of natural gas pipeline.

[0097] Among them, calling the solver to solve the mixed integer linear programming problem includes:

[0098] The prediction results are used as active load to solve for the specific value of .

[0099] As a preferred embodiment of the above, the economic benefits of each energy output under different optimization scheduling schemes are evaluated, including:

[0100] The economic benefits of different scheduling strategies include direct costs and indirect benefits. Direct costs include energy consumption, equipment wear and tear, and operation and maintenance costs; indirect benefits include energy utilization optimization and carbon emissions.

[0101] Combining the internal rate of return method, net present value analysis, sensitivity analysis and regression period analysis methods, the economic feasibility of different scheduling strategies is evaluated in multiple dimensions;

[0102] Game theory is introduced, and power generation companies randomly initialize their own bidding strategies within the upper and lower limits of the bidding strategies. The power generation and purchase costs, power generation profits, and energy utilization trend curves under various scenarios are obtained through multiple iterations.

[0103] The economic benefits and future development trends of each energy output under different scheduling strategies are obtained.

[0104] The PID control algorithm is used to adjust the output of gas-fired power generation and thermal power generation in real time, including:

[0105]

[0106] Where e(t) is the error signal of a single power generation system state, and the error signal is the error between the real-time output signal and the reference output signal; Δe(t) is the error signal change rate; K P , K I , KD are proportional, integral and differential coefficients; t is time.

[0107] like Figure 2 As shown, this embodiment also includes a control device for combined power supply of gas and thermal power generation, using the above method, the device includes:

[0108] A prediction unit is used to obtain historical electricity consumption data, and use a long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain a prediction result;

[0109] A modeling unit is used to optimize the output distribution of gas-fired power generation and thermal power generation using a mixed integer linear programming (MILP) method according to the prediction results to obtain a scheduling model;

[0110] The evaluation unit is used to set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme;

[0111] The control unit is used to monitor the changes in power load in real time, and use the PID control algorithm to adjust the output of gas-fired power generation and thermal power generation in real time based on the optimal scheduling plan.

[0112] Embodiment 2:

[0113] This embodiment includes a control strategy for combined power supply of gas-fired power generation and thermal power generation, including demand forecasting, resource scheduling, real-time control and economic analysis, including the following steps:

[0114] Demand forecasting uses big data and machine learning algorithms to predict electricity demand;

[0115] Resource scheduling optimizes the output distribution of gas-fired power generation and thermal power generation based on demand forecast results;

[0116] Real-time control: During actual operation, real-time monitoring of load changes in the power system is carried out, and the output of gas-fired power generation and thermal power generation is adjusted according to the changes;

[0117] Economic Analysis: Through economic analysis models, the economic feasibility of different scheduling strategies is evaluated and the optimal solution is selected.

[0118] like Figure 3 As shown, the demand forecast uses a long short-term memory (LSTM) neural network model to predict electricity demand, including:

[0119] The power consumption and total power consumption are normalized by using the normalization method with the minimum and maximum values ​​as the normalization standard. The corresponding formula is as follows:

[0120]

[0121] In the formula, P and W are the power consumption and total power consumption at the current moment, respectively. min and P max are the minimum and maximum values ​​of power consumption, W max and W min are the minimum and maximum values ​​of total power consumption, P user and W user are the normalized results respectively.

[0122] The normalized power consumption and total power consumption data are used as the input of the LSTM neural network model. Based on the LSTM unit structure and network connection structure, and after completing the setting of multiple different LSTM neural network hyperparameters, the prediction process of the LSTM neural network for power consumption and total power consumption is obtained. The calculation formula is as follows:

[0123] F=LSTM(epoch, batch-size, dropout) (3)

[0124] P′ user =F(P user ) (4)

[0125] W′ user =F(W user ) (5)

[0126] In the formula, F represents the LSTM neural network prediction model, epoch, batch-size and dropout are all network coefficients and need to be determined according to the network parameters, P′ user and W′ user are the power consumption and total power consumption calculated by the LSTM neural network prediction model, P user and W user are the normalized results respectively.

[0127] like Figure 4 As shown in the figure, the resource scheduling adopts the mixed integer linear programming (MILP) method to optimize the output allocation of gas-fired power generation and thermal power generation, including:

[0128] The optimization model includes two subsystems: the power system and the natural gas system. The gas-fired units are coupled to participate in the peak load regulation of power supply, and the output distribution of gas-fired power generation and thermal power generation is optimized to minimize the system operation cost.

[0129] Set the objective function C to minimize the running cost ope as follows:

[0130]

[0131] Among them, Cgas , C store , C gc , C g , C w , C ess They are the gas source output cost, gas storage tank operation cost, thermal power unit operation cost, wind power unit operation cost, gas turbine operation cost and power shortage and power restriction penalty cost; Ω s is a collection of typical day scenes; Ω gas is the natural gas source node set; Ω g is the gas turbine unit set; Ω c is a collection of coal-fired power generation units; store For gas storage equipment collection; Ω gen G is the collection of all generator sets; s，i，source is the total gas consumption of node i on a typical day; c i，gas is the contract gas purchase price at node i; c gc 、c w With c g They are the operation and maintenance costs per unit output of thermal power generating units, fans and gas turbines; c st，in 、c st，out They are the gas inlet and outlet costs per unit flow of the gas storage tank; Generate penalty costs for power shortage and power abandonment in the system; as well as They are the outputs of thermal power units, fans and gas turbines at time s and t on a typical day respectively; They are the gas inlet and outlet volumes of the gas storage tank respectively; The total output of each generator set in the system: is the active power consumed by the load at time s, t on a typical day.

[0132] Based on the objective function of minimizing operating costs, the optimization model sets constraints in multiple dimensions, including the balance of electricity and natural gas flow, equipment operation restrictions, and grid security constraints. It also performs convex optimization on the nonlinear terms involved and converts them into mixed integer linear programming problems. The power consumption in the prediction model is taken as the active load. The specific value of is further solved by the solver. The linearization of the Weymouth flow calculation equation of the natural gas subsystem is as follows:

[0133]

[0134]

[0135] in, is the tidal flow rate from m to n in the gas distribution network transmission pipeline at time t; and are the gas pressures of natural gas nodes m and n at time s and t on a typical day; K mn Weymouth characteristic parameters of natural gas pipeline.

[0136] A commercial solver is called to solve the mixed integer linear programming problem. Under the premise of satisfying all the given constraints, the optimal output allocation plan for gas-fired power generation and thermal power generation, as well as the corresponding start and stop plans for the generating units are determined.

[0137] like Figure 5 As shown, real-time control uses the SCADA system to monitor the changes in power load in real time, and uses the PID control algorithm to adjust the output of gas-fired power generation and thermal power generation in real time, including:

[0138] The analog signal of the power generation system collected in real time by the SCADA system is converted into a digital signal, and the output voltage of the integrator is detected by the zero-crossing comparator to control the logic control unit of the system. At the same time, the power generation output collected in real time by the SCADA system is used as the original data for load forecasting and optimal output solution, and the solution result is used as the basis for the start-stop decision of the PID control power generation system.

[0139] According to the optimal output of gas and thermal power generation systems, the output control quantity of gas and thermal power generation systems is adjusted through the PID control algorithm. Among them, the output control quantity of a single (gas or thermal) power generation system obtained by the PID control algorithm is as follows:

[0140]

[0141] Where e(t) is the error signal of a single power generation system state; Δe(t) is the error signal change rate; K P , K I , K D are proportional, integral and differential coefficients; t is time.

[0142] like Figure 6 As shown in the figure, economic analysis evaluates the economic feasibility of different scheduling strategies through cost-benefit analysis and selects the optimal solution, including:

[0143] The economic efficiency of different scheduling strategies is directly reflected in the two aspects of direct costs and indirect benefits. Direct costs include energy consumption, equipment wear and tear, and operation and maintenance costs, while indirect benefits include energy utilization optimization and carbon emissions. The economic efficiency of different scheduling strategies is further evaluated by combining internal rate of return method, net present value analysis, sensitivity analysis, and regression period analysis.

[0144] On the basis of the resource scheduling model, according to the energy development policy, different power generation scenarios are set for thermal power units, gas units and new energy output. Game theory is further introduced, and power generation companies randomly initialize their own bidding strategies within the upper and lower limits of the bidding strategy, and iterate multiple times to obtain the power generation and power purchase costs, power generation profits and energy utilization trend curves in each scenario. Based on this, combined with the long-term, multi-scenario scheduling optimization structure, the economic benefits and future development trends of each energy output under different scheduling strategies are obtained.

[0145] See also Figure 7 A computer device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores a computer program executable by the processor 410, and when the computer program is executed by the processor 410, the above method is executed.

[0146] The embodiment of the present application further provides a storage medium 430 on which a computer program is stored. When the computer program is run by the processor 410, the above method is executed.

[0147] Among them, the storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0148] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0149] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0150] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0151] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0152] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0153] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0154] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0155] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A control method for combined power supply of gas and thermal power generation, characterized in that: The steps include: Obtain historical electricity consumption data, and use a long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain a prediction result; According to the prediction results, the output distribution of gas-fired power generation and thermal power generation is optimized using the mixed integer linear programming MILP method to obtain a scheduling model; Set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme; The power load changes are monitored in real time, and the output of gas-fired power generation and thermal power generation is adjusted in real time using a PID control algorithm with reference to the optimal dispatching plan.

2. The control method for combined power supply of gas and thermal power generation according to claim 1, characterized in that: The method of using the long short-term memory (LSTM) neural network model to predict power demand includes the following steps: Normalizing the power consumption and total power consumption in the historical power consumption data; The normalized power consumption and total power consumption data are used as input of the LSTM neural network model; Based on the LSTM unit structure and network connection structure, complete the setting of multiple different LSTM neural network hyperparameters; Run the LSTM neural network model to predict power consumption and total power consumption.

3. The control method for combined power supply of gas and thermal power generation according to claim 2 is characterized in that: The normalizing of the power consumption and the total power consumption in the historical power consumption data includes: In the formula, P and W are the power consumption and total power consumption at the current moment, respectively. min and P max are the minimum and maximum values ​​of power consumption, W max and W min are the minimum and maximum values ​​of total power consumption, P user and W user are the normalized results respectively.

4. The control method for combined power supply of gas and thermal power generation according to claim 2 is characterized in that: The LSTM neural network model is run to predict the power consumption and total power consumption, and the calculation formula is as follows: F=LSTM(epoch,batch-size,dropout); P′ user =F(P user ); W′ user =F(W user ); Where F represents the LSTM neural network prediction model, epoch, batch-size and dropout are all network coefficients, which are determined according to the network parameters; P′ user and W′ user are the power consumption and total power consumption calculated by the LSTM neural network prediction model, P user and W user are the normalized results respectively.

5. The control method for combined power supply of gas and thermal power generation according to claim 1, characterized in that: The mixed integer linear programming MILP method is used to optimize the output distribution of gas power generation and thermal power generation to obtain a scheduling model, including: Taking the minimum operating cost as the objective function, a dispatching model is constructed. The dispatching model includes two subsystems, the power system and the natural gas system. The gas units in the natural gas system are coupled to participate in the power supply peak regulation, and the operating cost is minimized by optimizing the output distribution of gas-fired power generation and thermal power generation; Establish multi-dimensional constraints that take into account power and natural gas flow balance, equipment operation restrictions, and grid security; Convex optimization is performed on nonlinear terms in the scheduling model to convert it into a mixed integer linear programming problem; The solver is called to solve the mixed integer linear programming problem, and under the premise of satisfying the constraints, the optimal output allocation scheme of gas power generation and thermal power generation and the corresponding start and stop plan of the generator set are determined.

6. The control method for combined power supply of gas and thermal power generation according to claim 5, characterized in that: The objective function C ope include: In the formula, C gas , C store , C gc , C g , C w , C ess They are the gas source gas output cost, gas tank operation cost, thermal power unit operation cost, wind turbine unit operation cost, gas turbine operation cost and power shortage and power restriction penalty cost respectively; T is the scheduling optimization cycle, T = 24h; set is the total number of model scheduling scenarios; i, st, gc, g, w are the nodes where the gas source, gas tank, coal-fired unit, gas unit and wind turbine are located respectively; gen is the node where each generator set in the system is located; Ω s ,Ω gas ,Ω g ,Ω c ,Ω store and Ω gen They are typical day scenes, natural gas source nodes, gas turbine units, coal-fired generator units, gas storage equipment, and the collection of all generator unit equipment; G s,i,source G is the total gas consumption of node i on a typical day; s,i,source is the total gas consumption of node i on a typical day; c i,gas is the contract gas purchase price at node i; c gc 、c w With c g They are the operation and maintenance costs per unit output of thermal power generating units, fans and gas turbines; c st,in 、c st,out They are the gas inlet and outlet costs per unit flow of the gas storage tank; Generate penalty costs for power shortage and power abandonment in the system; as well as They are the outputs of thermal power units, fans and gas turbines at time s and t on a typical day respectively; They are the gas inlet and outlet volumes of the gas storage tank respectively; It is the sum of the output of each generator set in the system; is the active power consumed by the load at time s, t on a typical day.

7. The control method for combined power supply of gas and thermal power generation according to claim 5, characterized in that: The convex optimization of the nonlinear terms in the scheduling model includes: Linearize the flow calculation equations in the natural gas system: In the formula, is the tidal flow rate from m to n in the gas distribution network transmission pipeline at time t; and are the gas pressures of natural gas nodes m and n at time s and t on a typical day; K mn Weymouth characteristic parameters of natural gas pipeline.

8. The control method for combined power supply of gas and thermal power generation according to claim 6, characterized in that: The calling solver to solve the mixed integer linear programming problem includes: The prediction result is used as the active load to solve for the specific value of .

9. The control method for combined power supply of gas and thermal power generation according to claim 1, characterized in that: The evaluation of the economic benefits of each energy output under different optimization scheduling schemes includes: The economic benefits of different scheduling strategies include direct costs and indirect benefits. The direct costs include energy consumption, equipment wear and tear, and operation and maintenance costs; the indirect benefits include energy utilization optimization and carbon emissions. Combining the internal rate of return method, net present value analysis, sensitivity analysis and regression period analysis methods, the economic feasibility of different scheduling strategies is evaluated in multiple dimensions; Game theory is introduced, and power generation companies randomly initialize their own bidding strategies within the upper and lower limits of the bidding strategies. The power generation and purchase costs, power generation profits, and energy utilization trend curves under various scenarios are obtained through multiple iterations. The economic benefits and future development trends of each energy output under different scheduling strategies are obtained.

10. The control method for combined power supply of gas and thermal power generation according to claim 1, characterized in that: The PID control algorithm is used to adjust the output of gas power generation and thermal power generation in real time, including: Where, e(t) is the error signal of a single power generation system state, and the error signal is the error between the real-time output signal and the reference output signal; Δe(t) is the error signal change rate; K P , K I , K D are proportional, integral and differential coefficients; t is time.

11. A control device for combined power supply of gas and thermal power generation, characterized in that: Using the method according to any one of claims 1 to 10, the device comprises: A prediction unit, used to obtain historical electricity consumption data, and use a long short-term memory (LSTM) neural network model to predict electricity demand based on the historical electricity consumption data to obtain a prediction result; A modeling unit, configured to optimize the output distribution of gas-fired power generation and thermal power generation by using a mixed integer linear programming (MILP) method according to the prediction results to obtain a scheduling model; An evaluation unit is used to set different power generation scenarios, solve the optimal scheduling scheme of the scheduling model under long periods and multiple scenarios, evaluate the economic benefits of each energy output under different optimal scheduling schemes, and select the optimal scheduling scheme; The control unit is used to monitor the change of power load in real time, and to adjust the output of gas power generation and thermal power generation in real time by using PID control algorithm with reference to the optimal dispatching scheme.

12. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 10 is implemented.

13. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images