Energy-saving optimization method and device based on multi-unit heat supply system

Abstract

The invention discloses an energy-saving optimization method and device based on a multi-unit heat supply system, and belongs to the technical field of heat supply system optimization regulation, and the method comprises the following steps: S1, obtaining an operation data set of multiple units through a data sensor, and constructing a unit model of each unit; s2, a corrected generated power model and a corrected heat supply power model are obtained through data batch processing Spark; s3, establishing an optimization objective function; s4, based on an ant colony optimization algorithm, obtaining an optimal solution of the parameter set of the optimization objective function, and obtaining an optimization scheduling strategy through the optimal solution of the parameter set obtained by the unit model and the optimization objective function; and S5, optimizing the load distribution of the heat supply unit in real time. According to the energy-saving optimization method and device based on the multi-unit heat supply system, load distribution of the multiple heat supply units can be optimized in real time, and the operation efficiency of the multiple heat supply units of a thermal power plant is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optimization regulation of heating systems, and in particular to an energy-saving optimization method and device for a multi-unit heating system. Background Art

[0002] To address the environmental problems caused by heating, China is comprehensively promoting the transformation of energy production and consumption towards clean, low-carbon, and non-fossil methods. There have been significant changes in the energy supply method. In addition to using large and medium-sized cogeneration units and peak hot water boilers, efforts are being made to explore and promote the use of industrial waste heat, domestic waste, biomass energy, wind-powered electric boilers, and solar energy units as heating heat sources. In terms of the system operation goal, it is transitioning from a rough operation method mainly focused on meeting user needs to a refined regulation goal of being clean, low-carbon, and environmentally friendly.

[0003] Therefore, in terms of the regulation means for a multi-source heating system, more efficient and intelligent methods are needed to coordinately dispatch the operation loads of multiple heat sources for heating. In addition, with the rapid development of urbanization, the heating system is developing towards multi-city networking, multi-source complementarity, and interconnection, which further increases the complexity of the heat network. Especially the problem of load optimization dispatch between heat sources is a new problem for heating enterprises and a considerable challenge for enterprise operation dispatchers.

[0004] Currently, the operation strategies and control instructions of a multi-source heating system for thermal power generating units still mainly rely on manual experience and rules, lacking scientific data support and intelligent optimization methods, and it is difficult to achieve optimal operation. The operation dispatch method of heating enterprises on the heat source side is: taking the weather conditions as the condition, determining the total production load of the heating system, and then, according to manual experience, allocating the loads of different heat sources under the maximum heating capacity of the units and the transportation constraints on the heat network side. The manual dispatch method is simple, easy to implement, and highly operable, but it only considers the constraint conditions and does not consider the economy and environmental protection of operation dispatch, which will cause waste of operation costs. Secondly, in the process of the transformation of energy production towards cleaner methods, the coordinated operation dispatch of multiple heat sources will gradually exceed the decision-making ability of manual experience.

[0005] Therefore, there is an urgent need for an energy-saving optimization method and device for a multi-unit heating system. Summary of the Invention

[0006] The purpose of the present invention is to provide an energy-saving optimization method and device for a multi-unit heating system, which can optimize the load distribution of multiple heating units in real time and effectively improve the operation efficiency of multiple heating units in a thermal power plant.

[0007] To achieve the above purpose, the present invention provides an energy-saving optimization method for a multi-unit heating system, including the following steps:

[0008] Step S1: Obtain the operation dataset of multiple units through data sensors, and construct the unit models of each unit. The unit models of each unit include a power generation model and a heat supply model.

[0009] Step S2: Use data batch processing Spark to correct the pre-constructed power generation model and heat supply model respectively, and obtain the corrected power generation model and the corrected heat supply model.

[0010] Step S3: Based on the corrected power generation model, the corrected heat supply model, and the operation dataset, establish an optimization objective function.

[0011] Step S4: Based on the ant colony optimization algorithm, solve the optimization objective function to obtain the optimal solution of the parameter set, and obtain the optimal scheduling strategy through the optimal solution of the parameter set obtained by the unit model and the optimization objective function.

[0012] Step S5: Based on the optimal scheduling strategy, perform real-time optimization on the load distribution of heat supply units.

[0013] Preferably, step S1 includes the following sub-steps:

[0014] Step S101: Data acquisition:

[0015] Obtain the operation dataset D of multiple units through data sensors, where each sample contains a feature vector x i and a target value y i :

[0016] D = (x1, y1), (x2, y2), …, (x n , y n );

[0017] where i = 1, 2, 3, …, n;

[0018] Step S102: For each unit, use the linear regression algorithm to establish a model. Assuming there are k units, the model y ij of the jth unit is:

[0019] y ij = w j T x i + b j + ε ij ;

[0020] where w j represents the weight vector of the jth unit; w j T represents w j 's bias; b j represents the bias term of the jth unit; εij denotes the random error term.

[0021] Preferably, in step S2, the data batch processing Spark includes Transformations and Actions. Among them, Transformations is the data processing process, which uses the Spark framework to efficiently process large datasets. After batch processing, the corrected model parameters are and the corrected model is:

[0022]

[0023] where denotes the transpose of.

[0024] Preferably, in step S3, the optimization objective function is expressed as:

[0025]

[0026] where denotes the summation of j from 1 to k, and k is a real number greater than 1. is the cost function related to index j and and is specifically expressed as follows:

[0027]

[0028] where a j , b j , c j denote the coefficients related to the j-th unit; P(x) is the penalty term used to handle violations of the constraint conditions and is specifically expressed as follows:

[0029] P(x) = λ1P power (x) + λ2P heat (x) + λ3P capacity (x);

[0030] where λ1, λ2, and λ3 denote the weight coefficients used to adjust the importance of each sub-term; P power (x) denotes the penalty term for power demand constraints; P heat (x) denotes the penalty term for heating demand constraints; P capacity (x) denotes the penalty term for unit capacity limits.

[0031] Preferably, in step S4, it includes the update of pheromone and the calculation of the probability of ant path selection. The update of pheromone includes the evaporation process and the enhancement process;

[0032] The evaporation process is specifically as follows:

[0033] τ ij (t) = (1 - p)×τ ij (t - 1);

[0034] where τ ij (t) is the pheromone concentration on edge ij at time t of the event; p is the evaporation coefficient;

[0035] The enhancement process is specifically as follows:

[0036] τ′ ij (t) = τ ij (t) + Δτ ij ;

[0037] where τ′ ij (t) is the updated pheromone concentration on edge ij at time t; Δτ ij is the increased pheromone on path ij;

[0038]

[0039] where s is the number of parameters using this path in the current iteration; Q is a constant; L h is the path length that parameter h passes through.

[0040] Preferably, the probability calculation of ant path selection is specifically as follows:

[0041]

[0042] where is the probability that parameter h selects node j from node i; is the pheromone concentration on edge ij; is the heuristic information; J h is the set of adjacent nodes that parameter h can select; α and β are adjustment parameters, respectively controlling the influence degrees of pheromone and heuristic information;

[0043] After multiple iterations, the optimal parameter set x * is obtained, so that the optimization objective function F(x * ) reaches the minimum value.

[0044] Preferably, in step S5, based on the optimal parameter set x * obtained in step S4, combined with the actual operation data, the load distribution of each unit is dynamically adjusted to ensure that the system is always in the optimal operation state.

[0045] The present invention also provides an energy-saving optimization device based on a multi-unit heating system, including:

[0046] A data acquisition module, which acquires the operation data set of the multi-units;

[0047] A data processing module for establishing an optimization objective function;

[0048] A data analysis module for obtaining an optimized scheduling strategy;

[0049] A data optimization module for performing real-time optimization on the load distribution of a heat supply unit through the optimized scheduling strategy.

[0050] Preferably, a computer device includes: a memory and a processor; the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the energy-saving optimization method for a multi-unit heat supply system according to any one of claims 1-7 are implemented.

[0051] A computer-readable storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the energy-saving optimization method for a multi-unit heat supply system according to any one of claims 1-7 are implemented.

[0052] Therefore, by adopting the above energy-saving optimization method and device for a multi-unit heat supply system, the load distribution of multiple heat supply units can be optimized in real time, effectively improving the operation efficiency of multiple heat supply units in a thermal power plant.

[0053] Next, through the accompanying drawings and embodiments, the technical solutions of the present invention will be further described in detail. Description of the Drawings

[0054] Figure 1 is a flowchart of an embodiment of the energy-saving optimization method for a multi-unit heat supply system of the present invention;

[0055] Figure 2 is a structural diagram of an embodiment of the energy-saving optimization device for a multi-unit heat supply system of the present invention. Detailed Embodiments

[0056] The technical solutions of the present invention will be further described below through the accompanying drawings and embodiments.

[0057] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention belongs.

[0058] Embodiment 1

[0059] As Figure 1 shown, the present invention provides an energy-saving optimization method for a multi-unit heat supply system, including the following steps:

[0060] Step S1: Obtain the operation data sets of multiple units through data sensors, and construct the unit models of each unit. The unit models of each unit include a power generation model and a heat supply model.

[0061] Step S1 includes the following sub-steps:

[0062] Step S101: Data acquisition:

[0063] Obtain the operation data set D of multiple units through data sensors, where each sample contains a feature vector x i and a target value y i :

[0064] D = (x1, y1), (x2, y2), …, (x n , y n );

[0065] where i = 1, 2, 3, …, n;

[0066] Step S102: For each unit, use the linear regression algorithm to establish a model. Suppose there are k units, and the model y ij of the jth unit is:

[0067] y ij = w j T x i + b j + ε ij ;

[0068] where w j represents the weight vector of the jth unit; w j T represents the bias of w j ; b j represents the bias term of the jth unit; ε ij represents the random error term.

[0069] Step S2: Use data batch processing Spark to correct the pre-constructed power generation model and heat supply model respectively, and obtain the corrected power generation model and the corrected heat supply power model; the corrected power generation model is:

[0070] P a1 = η × μ1 × m a × LHV;

[0071] The corrected heat supply power model is:

[0072] P b1 = m b × [(h out × μ2) - h in ;

[0073] Among them, μ1 is the optimal solution of the thermal cycle efficiency obtained after data batch processing; μ2 is the optimal solution of the specific enthalpy at the outlet obtained after data batch processing.

[0074] Specifically, μ is:

[0075]

[0076] Among them, μ is the vector of the thermal cycle efficiency or the vector of the specific enthalpy at the outlet to be processed; Y K is the actual output data; is the output data of the model; L is the data length.

[0077] The data batch processing Spark includes Transformations and Actions. Among them, Transformations is the data processing process, including map, filter, and reduceBykey, and Actions are actions used to trigger calculations and return results.

[0078] After batch processing, the corrected model parameters are and The corrected model is:

[0079]

[0080] Among them, represents the transpose of.

[0081] Step S3, based on the corrected power generation model, the corrected heat supply power model, and the operation data set, establish an optimization objective function; aiming to minimize the total cost (such as fuel consumption, maintenance cost, etc.) and maximize the system performance (such as efficiency, stability). Consider multiple constraint conditions (such as power demand, heat supply demand, unit capacity limit, etc.).

[0082] The optimization objective function is expressed as:

[0083]

[0084] Among them, represents the sum of j from 1 to k, k is a real number greater than 1, is the cost function related to index j and and is specifically expressed as follows:

[0085]

[0086] Among them, a j 、b j 、cj Denote the coefficient related to the j-th unit; P(x) is the penalty term used to handle the violation of constraints, which is specifically expressed as follows:

[0087] P(x) = λ1P power (x) + λ2P heat (x) + λ3P capacity (x);

[0088] Among them, λ1, λ2, and λ3 represent the weight coefficients used to adjust the importance of each sub-term; P power (x) represents the penalty term for power demand constraints; P heat (x) represents the penalty term for heat supply demand constraints; P capacity (x) represents the penalty term for unit capacity limits.

[0089] Step S4: Based on the ant colony optimization algorithm, solve the optimization objective function to obtain the optimal solution of the parameter set, and obtain the optimal scheduling strategy through the optimal solution of the parameter set obtained by the unit model and the optimization objective function.

[0090] Ant Colony Optimization (ACO) is a heuristic search algorithm based on swarm intelligence, which is suitable for solving combinatorial optimization problems. Its core idea is to simulate the process of ants finding food paths to find the global optimal solution. The specific implementation includes the following aspects:

[0091] Initialization: Set the initial pheromone concentration, select the starting node, and set the maximum number of iterations.

[0092] Construct a solution: In each iteration, ants select the next node to visit according to the current pheromone concentration and heuristic information to form a path.

[0093] Update of pheromone and calculation of the probability of ant path selection. Adjust the pheromone concentration according to the quality of the solution found by the ants. The update of pheromone includes the evaporation process and the enhancement process;

[0094] The evaporation process is specifically as follows:

[0095] τ ij (t) = (1 - p) × τ ij (t - 1);

[0096] Among them, τ ij (t) is the pheromone concentration on the edge ij at time t of the event; p is the evaporation coefficient;

[0097] The enhancement process is specifically as follows:

[0098] τ′ ij (t) = τ ij (t) + Δτ ij ;

[0099] Among them, τ′ ij (t) is the updated pheromone concentration on edge ij at time t; Δτ ij is the increased pheromone on path ij;

[0100]

[0101] Among them, s is the number of parameters using this path in the current iteration; Q is a constant; L h is the path length passed by parameter h.

[0102] The probability calculation of ant path selection is specifically as follows:

[0103]

[0104] Among them, is the probability that parameter h selects node j from node i; is the pheromone concentration on edge ij; is the heuristic information; J h is the set of adjacent nodes that parameter h can select; α and β are adjustment parameters, respectively controlling the influence degrees of pheromone and heuristic information;

[0105] After multiple iterations, the optimal parameter set x * is obtained, so that the optimization objective function F(x * ) reaches the minimum value. At this time, the optimized scheduling strategy, that is, the best load distribution plan for each unit, can be obtained.

[0106] Step S5, based on the optimized scheduling strategy, perform real-time optimization on the load distribution of the heating units. Based on the optimal parameter set x * obtained in step S4, combined with the actual operation data, dynamically adjust the load distribution of each unit to ensure that the system is always in the optimal operation state. For example, according to the current power demand and heating demand, adjust the working mode of each unit (such as start / stop, increase / decrease load, etc.) to achieve the purpose of energy conservation and efficiency improvement.

[0107] An energy-saving optimization device based on a multi-unit heating system, as Figure 2 shown, includes:

[0108] A data acquisition module, which obtains the operation data set of multiple units;

[0109] A data processing module, which is used to establish an optimization objective function;

[0110] A data analysis module, which is used to obtain an optimized scheduling strategy;

[0111] A data optimization module, which performs real-time optimization on the load distribution of the heating units through the optimized scheduling strategy.

[0112] When the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0113] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0114] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memories), optical fiber devices, and portable compact disc read-only memories (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or processing it in other suitable ways when necessary, and then storing it in a computer memory.

[0115] Therefore, by adopting the above-mentioned energy-saving optimization method and device based on a multi-unit heating system, the present invention can optimize the load distribution of multiple heating units in real time and effectively improve the operating efficiency of multiple heating units in a thermal power plant.

[0116] It should be noted that the content not elaborated in detail in the present invention is prior art and well known to those skilled in the art.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions of the present invention or make equivalent replacements, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. An energy-saving optimization method based on a multi-unit heating system, characterized in that: The following steps are involved: Step S1, obtaining operation data sets of multiple units through data sensors, and constructing unit models of each unit, wherein the unit model of each unit includes a power generation model and a heat supply model; Step S2: respectively correct the pre-built power generation model and heating model through data batch processing Spark to obtain a corrected power generation model and a corrected heating power model; Step S3, establishing an optimization objective function based on the revised power generation model, the revised heating power model, and the operating data set; Step S4: Solve the optimization objective function based on the ant colony optimization algorithm to obtain the optimal solution of the parameter set, and obtain the optimal scheduling strategy through the optimal solution of the parameter set obtained by the unit model and the optimization objective function; Step S5: Based on the optimization scheduling strategy, the load distribution of the heating unit is optimized in real time.

2. The energy-saving optimization method based on a multi-unit heating system according to claim 1 is characterized in that: Step S1 includes the following sub-steps: Step S101: Data collection: The operating data set D of multiple units is obtained through data sensors, where each sample contains a feature vector x i and the target value y i : D=(x1,y1),(x2,y2),…,(x n ,y n ); Where i = 1, 2, 3, ..., n; Step S102: For each unit, a linear regression algorithm is used to establish a model. Assuming there are k units, the model y of the jth unit is ij for: y ij =w j T x i +b j +ε ij ; Among them, w j represents the weight vector of the jth unit; w j T Indicates w j The bias of j represents the bias term of the jth unit; ε ij represents the random error term.

3. The energy-saving optimization method based on a multi-unit heating system according to claim 2 is characterized in that: In step S2, the data batch processing Spark includes Transformations and Actions, where Transformations is the data processing process, using the Spark framework to efficiently process large data sets. After batch processing, the corrected model parameters are and Revised model for: in, express The transpose of .

4. The energy-saving optimization method based on a multi-unit heating system according to claim 3 is characterized in that: The optimization objective function in step S3 is expressed as: in, It means that j is summed from 1 to k, k is a real number greater than 1, is related to index j and The relevant cost function is specifically expressed as follows: Among them, a j 、b j 、c j represents the coefficient related to the jth unit; P(x) is the penalty term, which is used to deal with the violation of the constraint conditions, and is specifically expressed as follows: P(x)=λ1P power (x)+λ2P heat (x)+λ3P capacity (x); Among them, λ1, λ2, and λ3 represent weight coefficients, which are used to adjust the importance of each sub-item; P power (x) represents the penalty term of power demand constraint; P heat (x) represents the penalty term of the heating demand constraint; P capacity (x) represents the penalty term of unit capacity limitation.

5. The energy-saving optimization method based on a multi-unit heating system according to claim 4 is characterized in that: Step S4 includes the updating of pheromones and the probability calculation of ant path selection, and the updating of pheromones includes the volatilization process and the enhancement process; The specific volatilization process is: τ ij (t)=(1-p)×τ ij (t-1); Among them, τ ij (t) is the pheromone concentration on edge ij at event time t; p is the volatility coefficient; The specific enhancement process is as follows: t ij ′(t)=τ ij (t)+Δτ ij ; Among them, τ ij ′(t) is the updated pheromone concentration on edge ij at time t; Δτ ij is the pheromone added on path ij; Among them, s is the number of parameters using this path in the current iteration; Q is a constant; L h is the length of the path traversed by parameter h.

6. The energy-saving optimization method based on a multi-unit heating system according to claim 5 is characterized in that: The probability calculation of ant path selection is as follows: in, The probability of selecting node j from node i is parameter h; is the pheromone concentration on edge ij; is the heuristic information; J h is the set of adjacent nodes selected by parameter h; α and β are adjustment parameters, which control the influence of pheromone and heuristic information respectively; After multiple iterations, the optimal parameter set x is obtained * , so that the optimization objective function F(x * ) reaches its minimum value.

7. The energy-saving optimization method based on a multi-unit heating system according to claim 6 is characterized in that: In step S5, based on the optimal parameter set x obtained in step S4 * , dynamically adjust the load distribution of each unit to ensure that the system is always in the optimal operating state.

8. A device for energy-saving optimization method based on a multi-unit heating system according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, to obtain the operating data sets of multiple units; A data processing module is used to establish an optimization objective function; Data analysis module, used to obtain optimized scheduling strategies; The data optimization module optimizes the load distribution of heating units in real time by optimizing the scheduling strategy.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the energy-saving optimization method based on a multi-unit heating system according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the energy-saving optimization method based on a multi-unit heating system described in any one of claims 1 to 7 are implemented.

Images