Intelligent control method and device based on comprehensive target of biomass combustion heating system

Abstract

The application discloses a comprehensive target intelligent control method and device based on a biomass combustion heating system, which is characterized by the following steps: constructing operation parameters related to the biomass combustion heating system, performing online selection and preprocessing on operation data, modeling based on the processed data, and optimizing the parameter configuration model according to an asynchronous particle swarm optimization-genetic algorithm to obtain optimal parameter configurations and find out the optimal operation scheme, so that the characteristic indexes of the corresponding combustion state are optimized, the complex coupling relationship in the operation parameters of the biomass combustion heating system is solved, and the problem that optimal operation parameters are difficult to configure is solved; the optimal operation parameters are output by combining the asynchronous particle swarm optimization-genetic algorithm, the parameter configurations of the actual operation of the biomass combustion heating system are optimized, and thus the intelligent control of the biomass combustion heating system is realized.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control technology, and in particular relates to a comprehensive target intelligent control method and device based on a biomass combustion heating system. Background Art

[0002] As the only renewable carbon source, biomass has become a research hotspot due to its sustainable, abundant, and environmentally friendly characteristics. Biomass pyrolysis is a clean and efficient biomass conversion technology that converts biomass into clean-burning biofuels. This technology can not only be directly used in existing boilers and gas turbines, but can also be further refined and processed to produce liquid fuels with a quality approaching that of conventional fuels like diesel and gasoline.

[0003] Biomass heating is an important technology for energy conservation and emission reduction. Its goal is to achieve high efficiency and low emissions by adjusting the biomass fuel feed rate and the air distribution parameters of the biomass heating system. The combination of feed rate and air distribution parameters has a direct impact on the combustion state of the biomass heating system. Different configurations of these operating parameters, such as feed rate and air distribution, directly lead to different combustion efficiencies and pollutant emissions.

[0004] For a given biomass combustion heating system, under certain load conditions and for different combustion state characteristic indicators, there exists an optimal operating scheme that optimizes the corresponding combustion state characteristic indicators. However, the operating parameters of the biomass combustion heating system have a complex coupling relationship, and finding the optimal operating parameter configuration is not easy. Therefore, the combustion optimization of existing biomass combustion heating systems mainly relies on the experience of the staff, often only maintaining the combustion of the biomass furnace. How to optimize the configuration of the parameters in the actual operation of the biomass combustion heating system to achieve intelligent control of the biomass combustion heating system is a difficult problem that is difficult to solve with existing technologies. Summary of the Invention

[0005] In response to the above-mentioned deficiencies in the prior art, the present application provides a comprehensive target intelligent control method and device based on a biomass combustion heating system.

[0006] In the first aspect, the present application proposes a comprehensive target intelligent control method based on a biomass combustion heating system, comprising the following steps:

[0007] Construct the operation data of biomass combustion process based on biomass combustion heating system;

[0008] Select and preprocess the operating data, and build a parameter configuration model based on the preprocessed data;

[0009] Optimizing the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain an optimal parameter configuration;

[0010] Intelligent control of the biomass combustion heating system is achieved based on the optimal parameter configuration.

[0011] In some optional implementations of some embodiments, the biomass combustion-based heating system constructs operating data of the biomass combustion process, and the constructed operating data include: biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate.

[0012] In some optional implementations of some embodiments, selecting and preprocessing the operating data includes:

[0013] According to the combustion conditions of different biomass fuels, data selection is carried out in the operating data according to the principles of uniform distribution and equal quantity;

[0014] The selected data is preprocessed, and the data of each input quantity and the corresponding output quantity are in a state where the order of magnitude difference is less than 1 through unit transformation or multiplication coefficient method, and then the input quantity is normalized.

[0015] In some optional implementations of some embodiments, constructing the parameter configuration model based on the preprocessed data includes establishing the parameter configuration model using a support vector machine method;

[0016] The parameter configuration model:

[0017] in, is the feeding rate of biomass fuel, is the primary air distribution rate, is the secondary air distribution rate, and is the structural constant of the system.

[0018] In some optional implementations of some embodiments, optimizing the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain an optimal parameter configuration includes the following steps:

[0019] Step 1: Define the dimensional components of the particle swarm as the feed rate, primary air distribution rate, and secondary air distribution rate of the biomass fuel in the parameter configuration model;

[0020] Step 2: According to the actual operation requirements of the biomass combustion heating system, the optimal parameter configuration range of each operating parameter in the parameter configuration model is set, and the particle position, velocity, individual particle extreme value, asynchronous particle swarm optimization-genetic algorithm and genetic algorithm iteration number are initialized;

[0021] Step 3: Evaluate the fitness of all particles, including obtaining the fitness of particles and determining whether the particles are updated;

[0022] Step 4: Find the individual extreme value by comparing the fitness with the historical optimal individual particle;

[0023] Step 5: Obtain the local optimum by comparing the fitness with neighboring particles;

[0024] Step 6: Update particle velocity and position;

[0025] Step 7: Evaluate the fitness of the updated particles;

[0026] Step 8: Determine whether all particles in the ring have been updated. If any particles have not been updated, return to step 5.

[0027] Step 9: If all particles have been updated, select the best individual and perform the crossover operation in the genetic algorithm;

[0028] Step 10: After the crossover operation, select the particles with smaller fitness to retain:

[0029] Step 11: Determine whether to terminate by comparing the current number of iterations with the maximum number of iterations of the genetic algorithm;

[0030] Step 12: Determine whether the asynchronous particle swarm optimization-genetic algorithm is terminated. If it is terminated, input the optimal particles, that is, the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate, and use the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate as the optimal parameter configuration.

[0031] In the second aspect, the present application proposes a comprehensive target intelligent control device based on a biomass combustion heating system, comprising: an operation data construction module, a model construction module, a parameter optimization module and an intelligent control module;

[0032] The operation data construction module is used to construct the operation data of the biomass combustion process based on the biomass combustion heating system;

[0033] The model building module is used to select and preprocess the operating data and build a parameter configuration model based on the preprocessed data;

[0034] The parameter optimization module is used to optimize the parameter configuration model using asynchronous particle swarm optimization-genetic algorithm to obtain the optimal parameter configuration;

[0035] The intelligent control module is used to realize intelligent control of the biomass combustion heating system based on the optimal parameter configuration.

[0036] In some optional implementations of some embodiments, the operation data construction module includes a feed rate parameter construction unit, a first air distribution rate parameter construction unit, and a second air distribution rate parameter construction unit;

[0037] The feeding rate parameter construction unit is used to construct the feeding rate of the biomass fuel;

[0038] The first air distribution rate parameter construction unit is used to construct a primary air distribution rate;

[0039] The second air distribution rate parameter construction unit is used to construct a secondary air distribution rate.

[0040] In some optional implementations of some embodiments, the model building module includes a data processing unit;

[0041] The data processing unit is used to select data from the operating data according to the principles of uniform distribution and equal quantity for the combustion conditions of different biomass fuels, pre-process the selected data, and make the data of each input quantity and the corresponding output quantity have an order of magnitude difference of less than 1 through unit transformation or multiplication coefficient method, and then normalize the input quantity.

[0042] In some optional implementations of some embodiments, the model construction module further includes a parameter configuration model construction unit;

[0043] The parameter configuration model building unit is used to establish a parameter configuration model using a support vector machine method;

[0044] The parameter configuration model:

[0045] in, is the feeding rate of biomass fuel, is the primary air distribution rate, is the secondary air distribution rate, 、 and is the structural constant of the system.

[0046] In some optional implementations of some embodiments, the parameter optimization module includes a particle swarm definition unit, a particle initialization unit, a first particle fitness evaluation unit, an individual extreme value analysis unit, a local optimal analysis unit, a particle update unit, a second particle fitness evaluation unit, an update judgment unit, a crossover operation unit, a particle retention unit, a genetic algorithm iteration number judgment unit, and an optimal parameter configuration output unit;

[0047] The particle group defining unit is used for defining each dimension component of the particle group as a feeding rate of the biomass fuel, a primary air distribution rate and a secondary air distribution rate in the parameter configuration model respectively.

[0048] The particle initialization unit is used for setting an optimal parameter configuration range of each operation parameter in the parameter configuration model according to actual operation requirements of the biomass combustion heating system, and initializing particle positions, velocities, particle individual extremum, the asynchronous particle swarm optimization-genetic algorithm and the iteration number of the genetic algorithm.

[0049] The individual extremum analysis unit is used for finding the individual extremum through comparison with the fitness of the historical optimal individual particle.

[0050] The local optimum analysis unit is used for obtaining the local optimum through comparison with the fitness of adjacent particles.

[0051] The particle updating unit is used for updating the particle velocity and position.

[0052] The second particle fitness evaluation unit is used for evaluating the fitness of the updated particle.

[0053] The updating judgment unit is used for judging whether all particles in the ring have been updated, and if there is a particle that has not been updated, the local optimum analysis unit is executed.

[0054] The crossover operation unit is used for selecting the optimal individual to perform the crossover operation in the genetic algorithm if all particles have been updated.

[0055] The particle reservation unit is used for selecting the particle with smaller fitness to be reserved after the crossover operation.

[0056] The genetic algorithm iteration number judgment unit is used for judging whether to terminate through comparison of the current iteration number with the maximum iteration number of the genetic algorithm.

[0057] The optimal parameter configuration output unit is used for judging whether the asynchronous particle swarm optimization-genetic algorithm terminates, and if the asynchronous particle swarm optimization-genetic algorithm terminates, the optimal particle, i.e., the optimal feeding rate of the biomass fuel, the primary air distribution rate and the secondary air distribution rate, is inputted, and the optimal feeding rate of the biomass fuel, the primary air distribution rate and the secondary air distribution rate are taken as the optimal parameter configuration.

[0058] The present application has the following beneficial effects:

[0059] For biomass combustion heating systems, under certain load conditions, for different combustion state characteristic indicators, by constructing operating parameters related to the biomass combustion heating system, and performing online selection and preprocessing of the operating data, modeling is performed based on the processed data, and then the parameter configuration model is optimized according to the asynchronous particle swarm optimization-genetic algorithm to obtain the optimal parameter configuration, and the optimal operating scheme is found to optimize the characteristic indicators of the corresponding combustion state. This solves the complex coupling relationship in the operating parameters of the biomass combustion heating system and the problem that the optimal operating parameters are difficult to configure. Combined with the asynchronous particle swarm optimization-genetic algorithm to output the optimal operating parameters, the parameters in the actual operating biomass combustion heating system are optimized, thereby realizing intelligent control of the biomass combustion heating system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is the overall flow chart of the present invention.

[0061] Figure 2 Schematic diagram of the process of step 8.

[0062] Figure 3 Schematic diagram of the process of step 11.

[0063] Figure 4 Schematic diagram of the process of step 12.

[0064] Figure 5 This is a block diagram of the device principle of the present invention. DETAILED DESCRIPTION

[0065] The following describes exemplary embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein; rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0066] In the first aspect, this application proposes a comprehensive target intelligent control method based on a biomass combustion heating system, such as Figure 1 As shown, it includes steps S100-S400:

[0067] S100: constructing operation data of a biomass combustion process based on a biomass combustion heating system;

[0068] In some optional implementations of some embodiments, the biomass combustion-based heating system constructs operating data of the biomass combustion process, and the constructed operating data include: biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate.

[0069] Based on the actual operating conditions of the biomass combustion process, production costs include biomass fuel consumption, the power consumption of the feed system, and the primary and secondary fans. The variables that primarily determine heating costs are the biomass fuel feed rate, primary and secondary fan air distribution rates. During the combustion process of a biomass combustion heating system, the power consumption of the feed system is significantly smaller than that of the primary and secondary fans and can be ignored. Therefore, the construction of operational data primarily considers biomass fuel consumption and the power consumption of the primary and secondary fans. Therefore, the biomass fuel feed rate, primary and secondary air distribution rates serve as the primary operational data for subsequent calculations. This constructed operational data can be obtained from a real-time biomass combustion heating system operational database or directly collected through instrumentation. The variety of biomass fuels used in biomass combustion heating systems is generally limited (due to different sources and industrial analytical data). Therefore, separate data collection is required for different biomass fuels (with different raw materials and sources) to facilitate targeted modeling.

[0070] S200: Select and preprocess the operating data, and build a parameter configuration model based on the preprocessed data;

[0071] In some optional implementations of some embodiments, selecting and preprocessing the operating data includes:

[0072] According to the combustion conditions of different biomass fuels, data selection is carried out in the operating data according to the principles of uniform distribution and equal quantity;

[0073] The selected data is preprocessed, and the data of each input quantity and the corresponding output quantity are in a state where the order of magnitude difference is less than 1 through unit transformation or multiplication coefficient method, and then the input quantity is normalized.

[0074] Among them, due to the limited variation of biomass fuels, data selection is carried out according to the combustion conditions of different fuels, and sample data for modeling is selected. The following principles are followed when selecting:

[0075] ① Uniform distribution, that is, the distribution of the input quantity of the model is uniform in the topological structure space; ② Equal quantity, that is, the sample size of the data at different points in the topological structure does not differ by more than 10% of the sample size of the point with the least data;

[0076] In some optional implementations of some embodiments, constructing the parameter configuration model based on the preprocessed data includes establishing the parameter configuration model using a support vector machine method;

[0077] The parameter configuration model:

[0078] in, is the feeding rate of biomass fuel, is the primary air distribution rate, is the secondary air distribution rate, 、 and is the structural constant of the system.

[0079] S300: Optimizing the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain an optimal parameter configuration;

[0080] In some optional implementations of some embodiments, optimizing the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain an optimal parameter configuration includes the following steps:

[0081] Step 1: Define the dimensional components of the particle swarm as the feed rate, primary air distribution rate, and secondary air distribution rate of the biomass fuel in the parameter configuration model;

[0082] Step 2: According to the actual operation requirements of the biomass combustion heating system, the optimal parameter configuration range of each operating parameter in the parameter configuration model is set, and the particle position, velocity, individual particle extreme value, asynchronous particle swarm optimization-genetic algorithm and genetic algorithm iteration number are initialized;

[0083] Furthermore, initialize the particle position and velocity:

[0084] Initialize individual particle extrema:

[0085] Initialize the number of iterations of the asynchronous particle swarm optimization-genetic algorithm (hereinafter referred to as APSO-GA algorithm) and the genetic algorithm (hereinafter referred to as GA algorithm):

[0086] Among them, i represents the number of particles, which is the i-th particle; j represents the particle dimension number, which is the j-th dimension vector of the particle, and j represents 、 and , that is, j can be defined as all parameters of biomass fuel feed rate, primary air distribution rate and secondary air distribution rate; rand(1) represents a random number uniformly distributed between [0, 1]; P besti Indicates the individual optimality of the i-th particle; I terP Indicates the number of iterations of the APSO-GA algorithm; I terG Indicates the number of iterations of the GA algorithm;

[0087] Step 3: Evaluate the fitness of all particles, including obtaining the fitness of particles and determining whether the particles are updated;

[0088] Further, the fitness of the particle needs to be obtained first:

[0089] wherein f is a fitness function used to quantitatively evaluate the given particle, which is related to the function to be implemented, and is used in the present embodiment to represent the calculation of the optimal parameter configuration (i.e. the optimal solution), represents the dimension of the particle, i.e. the number of independent variables, in the present embodiment, because j represents , and i.e. j can be defined as all parameters of the feeding rate of the biomass fuel, the primary air distribution rate and the secondary air distribution rate, includes the three independent variables of the feeding rate of the biomass fuel, the primary air distribution rate and the secondary air distribution rate, so ;

[0090] Initialize the number of particle updates in the APSO loop to determine whether all particles in the loop have been updated.

[0091] , nu represents the number of particle updates in the APSO particle loop, which is used to determine whether all particles have been updated;

[0092] Step 4: Find the individual extremum by comparing the fitness of the historical optimal individual particle;

[0093] Further, find the individual maximum (or minimum) value of all particles by comparing the fitness of the historical individual optimum, the selection of maximum or minimum is related to the setting of the fitness function and the function to be implemented. Hereinafter, the minimum value is taken as an example:

[0094] wherein the initial value of the particle serial number i is set to 1, in preparation for particle updating, , represents the position of the jth dimension vector of the ith particle in the Ith iteration, terP represents the velocity of the jth dimension vector of the ith particle in the Ith iteration; terP

[0095] Step 5: Obtain the local optimum by comparing the fitness of adjacent particles;

[0096] Compare the fitness of the ith particle with that of its adjacent particles to obtain the local optimum. Wherein represents the individual extremum of the updated particle, then: ​​

[0097] in, is the neighborhood of particle i, and the algorithm takes one particle on the left and one particle on the right as its neighborhood; L besti represents the local optimum of the i-th particle;

[0098] Step 6: Update particle velocity and position;

[0099] Update speed:

[0100] Update location:

[0101] in (1), is a random number uniformly distributed between [0, 1]; c1 and c2 dimensional acceleration coefficients; is the inertia factor, which is used to update the particle velocity. The formula is as follows:

[0102] Update particle update times: nu=nu+1, I terPmax Indicates the maximum number of algorithm iterations, used to terminate the iteration, and the value range is [100,4000]; Indicates the minimum value of the inertia factor, which is 0.4; Indicates the maximum value of the inertia factor, which is 0.9. The inertia factor indicates the influence of the velocity of the previous generation of particles on the velocity of the current generation of particles. It is beneficial to global search, jump out of local extreme values, and avoid falling into local optimality; while smaller It is conducive to local search and allows the algorithm to quickly converge to the optimal solution.

[0103] Step 7: Evaluate the fitness of the updated particles;

[0104] Among them, the fitness calculation is performed on the i-th particle after the update, and the optimal individual particle is updated and used to update the next particle.

[0105] Step 8: Determine whether all particles in the ring have been updated. If any particles have not been updated, return to step 5.

[0106] Among them, such as Figure 2 As shown in the figure, the particles in the algorithm are arranged in a ring. If there are particles that have not been updated, return to step 5.

[0107] Step 9: If all particles have been updated, select the best individual and perform the crossover operation in the genetic algorithm;

[0108] If all particles in the ring have been updated, the best individual is selected and the crossover operation in the genetic algorithm is performed. Assume that the position of the selected particle is x i ’ , choose the number of individuals as M, then:

[0109] In order to select individuals with smaller fitness, the bubble method is used to sort all particles in the population P in ascending fitness order:

[0110] Where q represents the number of bubble sort traversals, that is, the number of cycles of the inertia factor formula in step 6. The inertia factor formula in step 6 needs to be cycled P-1 times to complete the sorting of all particles. The selected individuals are the particles with individual numbers 1 to M after the cycle P-1 times; x * Represents position auxiliary parameters, used for particle sorting.

[0111] Select M individuals:

[0112] Among them, x i ’ represents the position of the particle selected in step 9, which is used for the refinement of the genetic algorithm; x i q It represents the position of the i-th particle in the sorting after the q-th traversal; M represents the number of individual particles selected for refinement.

[0113] Step 10: After the crossover operation, select the particles with smaller fitness to retain:

[0114] After the crossover operation, particles with smaller fitness are selected and retained for the next iteration:

[0115] Crossover operation:

[0116] Among them, x i(i+1) is the daughter particle generated by the intersection,

[0117] Particle Selection:

[0118] Step 11: Determine whether to terminate by comparing the current number of iterations with the maximum number of iterations of the genetic algorithm;

[0119] Among them, such as Figure 3 As shown, the termination is determined by comparing the current number of iterations with the maximum number of GA iterations. terGmax ---The maximum number of genetic algorithm iterations, used to terminate the genetic algorithm iteration, is 6.

[0120] Step 12: Determine whether the asynchronous particle swarm optimization-genetic algorithm is terminated. If it is terminated, input the optimal particles, that is, the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate, and use the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate as the optimal parameter configuration.

[0121] Among them, the optimized particles are put back into the remaining particles of the population P and reordered for use in the next iteration of APSO:

[0122] like Figure 4 As shown in the figure, the termination is determined by comparing the current number of iterations with the maximum number of iterations of APSO-GA: the optimal x finally output is the optimal particle, that is, the solution of the optimal dimensions of the feeding rate, primary air distribution rate and secondary air distribution rate of the biomass fuel, indicating that the feeding rate, primary air distribution rate and secondary air distribution rate of the biomass fuel are configured as the optimal parameters.

[0123] S400: Implementing intelligent control of the biomass combustion heating system based on the optimal parameter configuration.

[0124] The optimal solutions of the biomass fuel feed rate, primary air distribution rate and secondary air distribution rate are input into the parameter configuration model to obtain the optimal combination of adjustable operating parameters, thereby realizing the optimal parameter configuration of the biomass combustion heating system and obtaining the optimal operation plan, so as to optimize the characteristic indicators of the corresponding combustion state, solve the complex coupling relationship in the operating parameters of the biomass combustion heating system, and solve the problem that the optimal operating parameters are difficult to configure, thereby realizing the intelligent control of the biomass combustion heating system.

[0125] In the second aspect, the present application proposes a comprehensive target intelligent control device based on a biomass combustion heating system, comprising: an operation data construction module, a model construction module, a parameter optimization module and an intelligent control module;

[0126] The operation data construction module is used to construct the operation data of the biomass combustion process based on the biomass combustion heating system;

[0127] The model building module is used to select and preprocess the operating data and build a parameter configuration model based on the preprocessed data;

[0128] The parameter optimization module is used to optimize the parameter configuration model using asynchronous particle swarm optimization-genetic algorithm to obtain the optimal parameter configuration;

[0129] The intelligent control module is used to realize intelligent control of the biomass combustion heating system based on the optimal parameter configuration.

[0130] In some optional implementations of some embodiments, the operation data construction module comprises a feeding rate parameter construction unit, a first air distribution rate parameter construction unit and a second air distribution rate parameter construction unit;

[0131] The feeding rate parameter construction unit is configured to construct a feeding rate of the biomass fuel.

[0132] The first air distribution rate parameter construction unit is configured to construct a primary air distribution rate.

[0133] The second air distribution rate parameter construction unit is configured to construct a secondary air distribution rate.

[0134] In some optional implementations of some embodiments, the model construction module comprises a data processing unit.

[0135] The data processing unit is configured to, for combustion of different biomass fuels respectively, select data in operation data according to a distribution uniformity and quantity equality principle, pre-process the selected data, and through unit conversion or multiplication coefficient method, make data of each input and corresponding output in a state that a quantity order difference is less than 1, and then perform normalization processing on the input.

[0136] In some optional implementations of some embodiments, the model construction module further comprises a parameter configuration model construction unit.

[0137] The parameter configuration model construction unit is configured to establish a parameter configuration model by using a support vector machine method.

[0138] The parameter configuration model is:

[0139] wherein, is a feeding rate of the biomass fuel, is a primary air distribution rate, is a secondary air distribution rate, , and is a structural constant of the system.

[0140] In some optional implementations of some embodiments, the parameter optimization module comprises a particle swarm definition unit, a particle initialization unit, a first particle fitness evaluation unit, an individual extremum analysis unit, a local optimum analysis unit, a particle updating unit, a second particle fitness evaluation unit, an updating judgment unit, a crossover operation unit, a particle reservation unit, a genetic algorithm iteration number judgment unit and a best parameter configuration output unit.

[0141] The particle swarm definition unit is used to define the dimensional components of the particle swarm, which are the feed rate, primary air distribution rate and secondary air distribution rate of the biomass fuel in the parameter configuration model;

[0142] The particle initialization unit is used to set the optimal parameter configuration range of each operating parameter in the parameter configuration model according to the actual operating requirements of the biomass combustion heating system, and initialize the particle position, velocity, individual particle extreme value, asynchronous particle swarm optimization-genetic algorithm and genetic algorithm iteration number; the first particle fitness evaluation unit is used to evaluate the fitness of all particles, including obtaining particle fitness and determining whether the particles are updated;

[0143] The individual extreme value analysis unit is used to find the individual extreme value by comparing the fitness with the historical optimal individual particle;

[0144] The local optimum analysis unit is used to obtain the local optimum by comparing the fitness with adjacent particles;

[0145] The particle updating unit is used to update the particle speed and position;

[0146] The second particle fitness evaluation unit is used to evaluate the fitness of the updated particles;

[0147] The update judgment unit is used to judge whether all particles in the ring have been updated. If there are particles that have not been updated, it returns to the local optimal analysis unit for execution;

[0148] The crossover operation unit is used to select the best individual and perform a crossover operation in the genetic algorithm if all particles have been updated;

[0149] The particle retention unit is used to select particles with smaller fitness to retain after performing the crossover operation:

[0150] The genetic algorithm iteration number judgment unit is used to judge whether to terminate by comparing the current iteration number with the maximum iteration number of the genetic algorithm;

[0151] The optimal parameter configuration output unit is used to determine whether the asynchronous particle swarm optimization-genetic algorithm is terminated. If terminated, the optimal particles, that is, the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate are input, and the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate are used as the optimal parameter configuration.

[0152] The above are only preferred embodiments of the present invention. It should be pointed out that various modifications and improvements made by those skilled in the art without departing from the present technical solution should also be deemed to fall within the scope of protection required by the claims.

Claims

1. A comprehensive target intelligent control method for a biomass combustion heating system, characterized by: The following steps are involved: Construct the operation data of biomass combustion process based on biomass combustion heating system; Select and pre-process the operating data, including: selecting data based on the principles of uniform distribution and equal quantity in the operating data according to the combustion conditions of different biomass fuels; Preprocessing the selected data, making the data of each input quantity and the corresponding output quantity less than 1 order of magnitude difference by unit transformation or coefficient multiplication, then normalizing the input quantity, and constructing a parameter configuration model based on the preprocessed data, including using a support vector machine method to establish the parameter configuration model; The parameter configuration model: in, is the feeding rate of biomass fuel, is the primary air distribution rate, is the secondary air distribution rate, 、 and is the structural constant of the system; Optimizing the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain the optimal parameter configuration includes the following steps: Step 1: Define the dimensional components of the particle swarm as the feed rate, primary air distribution rate, and secondary air distribution rate of the biomass fuel in the parameter configuration model; Step 2: According to the actual operation requirements of the biomass combustion heating system, the optimal parameter configuration range of each operating parameter in the parameter configuration model is set, and the particle position, velocity, individual particle extreme value, asynchronous particle swarm optimization-genetic algorithm and genetic algorithm iteration number are initialized; Step 3: Evaluate the fitness of all particles, including obtaining the fitness of particles and determining whether the particles are updated; Step 4: Find the individual extreme value by comparing the fitness with the historical optimal individual particle; Step 5: Obtain the local optimum by comparing the fitness with neighboring particles; Step 6: Update particle velocity and position; Step 7: Evaluate the fitness of the updated particles; Step 8: Determine whether all particles in the ring have been updated. If any particles have not been updated, return to step 5. Step 9: If all particles have been updated, select the best individual and perform the crossover operation in the genetic algorithm; Step 10: After the crossover operation, select the particles with smaller fitness to retain: Step 11: Determine whether to terminate by comparing the current number of iterations with the maximum number of iterations of the genetic algorithm; Step 12: Determine whether the asynchronous particle swarm optimization-genetic algorithm is terminated. If it is terminated, input the optimal particles, that is, the optimal biomass fuel feeding rate, primary air distribution rate, and secondary air distribution rate, and use the optimal biomass fuel feeding rate, primary air distribution rate, and secondary air distribution rate as the optimal parameter configuration; Intelligent control of the biomass combustion heating system is achieved based on the optimal parameter configuration.

2. The method according to claim 1, wherein: The biomass combustion-based heating system constructs operating data of the biomass combustion process, and the constructed operating data includes: a feeding rate of the biomass fuel, a primary air distribution rate, and a secondary air distribution rate.

3. The comprehensive target intelligent control device based on the biomass combustion heating system is characterized by: include: Run the data construction module, model building module, parameter optimization module and intelligent control module; The operation data construction module is used to construct the operation data of the biomass combustion process based on the biomass combustion heating system; The model construction module includes a data processing unit and a parameter configuration model construction unit; The data processing unit is used to select data from the operating data according to the principles of uniform distribution and equal quantity for the combustion conditions of different biomass fuels, pre-process the selected data, and make the data of each input quantity and the corresponding output quantity have an order of magnitude difference of less than 1 by unit conversion or coefficient multiplication. The input quantity is then normalized for selection and pre-processing of the operating data, and a parameter configuration model is constructed based on the pre-processed data. The parameter configuration model building unit is used to establish a parameter configuration model using a support vector machine method; in, is the feeding rate of biomass fuel, is the primary air distribution rate, is the secondary air distribution rate, 、 and is the structural constant of the system; The parameter optimization module is used to optimize the parameter configuration model using an asynchronous particle swarm optimization-genetic algorithm to obtain the optimal parameter configuration. The parameter optimization module includes a particle swarm definition unit, a particle initialization unit, a first particle fitness evaluation unit, an individual extreme value analysis unit, a local optimal analysis unit, a particle update unit, a second particle fitness evaluation unit, an update judgment unit, a crossover operation unit, a particle retention unit, a genetic algorithm iteration number judgment unit, and an optimal parameter configuration output unit; The particle swarm definition unit is used to define the dimensional components of the particle swarm, which are the feed rate, primary air distribution rate and secondary air distribution rate of the biomass fuel in the parameter configuration model; The particle initialization unit is used to set the optimal parameter configuration range of each operating parameter in the parameter configuration model according to the actual operating requirements of the biomass combustion heating system, and initialize the particle position, velocity, individual particle extreme value, asynchronous particle swarm optimization-genetic algorithm and genetic algorithm iteration number; the first particle fitness evaluation unit is used to evaluate the fitness of all particles, including obtaining particle fitness and determining whether the particles are updated; The individual extreme value analysis unit is used to find the individual extreme value by comparing the fitness with the historical optimal individual particle; The local optimum analysis unit is used to obtain the local optimum by comparing the fitness with adjacent particles; The particle updating unit is used to update the particle speed and position; The second particle fitness evaluation unit is used to evaluate the fitness of the updated particles; The update judgment unit is used to judge whether all particles in the ring have been updated. If there are particles that have not been updated, it returns to the local optimal analysis unit for execution; The crossover operation unit is used to select the best individual and perform a crossover operation in the genetic algorithm if all particles have been updated; The particle retention unit is used to select particles with smaller fitness to retain after performing the crossover operation: The genetic algorithm iteration number judgment unit is used to judge whether to terminate by comparing the current iteration number with the maximum iteration number of the genetic algorithm; The optimal parameter configuration output unit is used to determine whether the asynchronous particle swarm optimization-genetic algorithm is terminated. If it is terminated, the optimal particles, that is, the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate are input, and the optimal biomass fuel feeding rate, primary air distribution rate and secondary air distribution rate are used as the optimal parameter configuration; The intelligent control module is used to realize intelligent control of the biomass combustion heating system based on the optimal parameter configuration.

4. The device according to claim 3, characterized in that: The operation data construction module includes a feed rate parameter construction unit, a first air distribution rate parameter construction unit and a second air distribution rate parameter construction unit; The feeding rate parameter construction unit is used to construct the feeding rate of the biomass fuel; The first air distribution rate parameter construction unit is used to construct a primary air distribution rate; The second air distribution rate parameter construction unit is used to construct a secondary air distribution rate.

Images