Intelligent energy-saving control method and device for gas boiler steam-water system based on fuzzy control

Abstract

The application discloses a gas boiler steam-water system intelligent energy-saving regulation and control method and device based on fuzzy control, relates to the technical field of intelligent control of gas boilers, and acquires historical operation data and real-time operation data, carries out fuzzy processing on the data by using a quantum state superposition model, establishes a four-dimensional decision matrix as a time-space correlation controller, dynamically adjusts decision weights by using a Lyapunov index, generates a field effect control law based on flow field vortex characteristics to adaptively adjust combustion intensity, starts an abnormal propagation blocking algorithm when detecting local overheating abnormality, suppresses overheating by reverse propagation of a control equation, and simultaneously calls meta-evolution fuzzy rules to realize self-organization optimization through environment perception, rule competition and structure reconstruction. The method can improve the operation efficiency and energy-saving effect of the gas boiler and guarantee safe operation of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control of gas boilers, more particularly, to an intelligent energy-saving regulation and control method and device for a gas boiler steam-water system based on fuzzy control. BACKGROUND

[0002] The gas boiler steam-water system is a key equipment in industrial production, and its operation efficiency and energy consumption level are crucial to production cost and environmental impact. Traditional regulation and control methods mainly rely on manual experience or simple automatic control systems. These methods have many limitations when facing complex working condition changes. First, traditional methods are difficult to handle the uncertainty and fuzziness of system operation data, and cannot effectively utilize historical data for accurate prediction and optimization. Second, traditional fuzzy control systems are usually based on plane fuzzy rules and lack consideration of spatiotemporal correlation characteristics, making it difficult to adapt to complex combustion processes. In addition, existing technologies also have deficiencies in anomaly detection and handling, which cannot timely detect and suppress local overheating and other abnormal conditions, which may cause equipment damage and safety hazards. With the development of artificial intelligence, quantum computing, and complex system control theory, the combination of quantum state superposition model and fuzzy control can realize parallel fuzzy processing of multiple parameters by using the mapping of quantum probability amplitude and quantum entanglement characteristics, which can more accurately reflect the real operation state of the system. Meanwhile, the introduction of a four-dimensional decision matrix as a spatiotemporal correlation controller, combined with the dynamic weight migration algorithm of Lyapunov index, can adjust the decision weight in real time, further improving the adaptability and stability of the system. In addition, through holographic physical field perception driven control and meta-evolution fuzzy rules, anomaly detection, blocking, and dynamic optimization of rules can be realized, providing a new solution for intelligent energy-saving regulation and control of the gas boiler steam-water system, which can improve combustion efficiency, reduce energy consumption, and ensure safe and stable operation of the system.

[0003] Therefore, the existing technology has deficiencies in handling data fuzziness, spatiotemporal correlation, and anomaly handling, making it difficult to achieve accurate regulation and energy optimization. SUMMARY

[0004] To overcome the deficiencies of existing technologies in handling data fuzziness, spatiotemporal correlation, and anomaly handling, and to solve the problem of difficulty in achieving accurate regulation and energy optimization, the present application discloses an intelligent energy-saving regulation and control method and device for a gas boiler steam-water system based on fuzzy control, which can effectively solve the above technical problems.

[0005] To solve the above technical problems, the technical solution of the present application is as follows:

[0006] The intelligent energy-saving regulation and control method for a gas boiler steam-water system based on fuzzy control comprises the following steps:

[0007] Acquire historical and real-time operating data sets of the gas-fired boiler steam-water system. The historical operating data set includes combustion efficiency, temperature field distribution, vibration frequency, and energy consumption data at different time periods. The real-time operating data set includes current flow field vortex characteristics, acoustic emission signals, infrared thermal image data, and thermal parameters.

[0008] The real-time running data set is input into the quantum state superposition model |μ>=α|Low>+β|Medium>+γ|High> to perform fuzzification processing on the data, where |μ> represents the fuzzified data state, |Low>, |Medium>, and |High> represent the low, medium, and high fuzzy sets, respectively, and α, β, and γ are quantum probability amplitudes satisfying |α| 2 +|β| 2 +|γ| 2 =1;

[0009] A four-dimensional decision matrix is ​​established as a spatiotemporal correlation controller. The four-dimensional decision matrix uses time and space three-dimensional coordinates as dimensions to replace the traditional planar fuzzy rules.

[0010] Based on the Lyapunov index, the decision weights of each dimension of the four-dimensional decision matrix are adjusted in real time using a dynamic weight transfer algorithm. The weight adjustment formula is as follows:

[0011] ω i =exp(λ i ) / ∑exp(λ j )

[0012] Where λ i Let ω be the Lyapunov exponent of the i-th dimension. i These are the decision weights for the corresponding dimensions;

[0013] Combustion intensity is adaptively adjusted by generating a field-effect control law based on the vortex characteristics of the flow field. The formula for the field-effect control law is as follows:

[0014]

[0015] Where Ω is the vortex intensity. For the temperature gradient, k1, k2, and k3 are proportionality coefficients;

[0016] When a local overheating anomaly is detected, the anomaly propagation blocking algorithm is activated to suppress the local overheating in a timely manner through backpropagation of the control equation, which is:

[0017] Where v is the fluid velocity, α is the thermal diffusivity, β is the blocking coefficient, δ is the Dirac function, and r0 is the location of the anomaly source;

[0018] The meta-evolutionary fuzzy rules are invoked, and the meta-evolutionary fuzzy rules construct a deformable rule unit library, whose runtime self-organizing mechanism includes:

[0019] Environmental perception layer: fuses data from multiple sources of sensors, including vibration sensors, acoustic emission sensors, and infrared sensors;

[0020] Evolution-driven layer: A game-theoretic rule-based competition and elimination mechanism filters rule units;

[0021] Structural Reconstruction Layer: Dynamically adjusts the rule structure through a topological gene recombination algorithm;

[0022] Based on the above processing, a load resource regulation scheme is generated to carry out intelligent energy-saving regulation of the gas boiler steam-water system.

[0023] Preferably, the quantum state superposition model performs fuzzification processing on the data, including:

[0024] Map each parameter in the real-time running data set to the quantum state space, and determine the quantum probability amplitudes α, β, and γ corresponding to each parameter;

[0025] By leveraging the properties of quantum entanglement, parallel fuzzification of multiple parameters is achieved, generating quantum fuzzy sets.

[0026] Preferably, establishing a four-dimensional decision matrix as a spatiotemporal correlation controller includes:

[0027] The matrix is ​​defined by time t and spatial coordinates x, y, and z.

[0028] Each matrix element corresponds to a control rule for a specific spatiotemporal point, and its value is determined based on the parameter correlation degree of that spatiotemporal point.

[0029] Preferably, the step of generating a field effect control law based on the vortex characteristics of the flow field to adaptively adjust the combustion intensity includes:

[0030] A flow field model is constructed using computational fluid dynamics, and characteristic parameters of the flow field vortex are extracted.

[0031] Based on the mapping relationship between vortex characteristic parameters and combustion efficiency, the parameters of the field effect control law are dynamically adjusted.

[0032] Preferably, the step of activating the anomaly propagation blocking algorithm when a local overheating anomaly is detected, and timely suppressing local overheating through backpropagation of the control equation, includes:

[0033] The location and temperature of localized overheated areas can be identified using infrared thermal imaging data.

[0034] Solving the control equation yields the optimal value of the blocking coefficient β, and the control actuator injects a reverse energy flow to suppress overheat propagation.

[0035] Preferably, the structural reconstruction layer: dynamically adjusting the rule structure through a topological gene recombination algorithm includes:

[0036] The topological structure of regular units is encoded to form regular genes;

[0037] Genetic algorithms are used to perform crossover and mutation operations on regular genes to generate new regular structures.

[0038] The new rule structure is evaluated based on its control effect, and the optimal rule structure is retained.

[0039] Preferably, the intelligent energy-saving control device for a gas-fired boiler steam-water system based on fuzzy control includes:

[0040] Data acquisition module: used to acquire historical and real-time operating data sets of the gas-fired boiler steam-water system;

[0041] The spatiotemporally coupled quantum fuzzy control module is used to fuzzify data using a quantum state superposition model, establish a four-dimensional decision matrix, and adjust the decision weights of each dimension based on the Lyapunov exponent using a dynamic weight transfer algorithm.

[0042] Holographic physical field perception and drive control module: used to generate field effect control law based on flow field vortex characteristics, and to start an anomaly propagation blocking algorithm when a local overheating anomaly is detected;

[0043] Meta-evolutionary fuzzy rule module: includes an environment perception layer, an evolutionary driving layer and a structure reconstruction layer, used to build a deformable rule unit library and achieve runtime self-organization;

[0044] Control and execution module: Used to control the steam and water system of the gas boiler according to the generated load resource control scheme.

[0045] Preferably, the spatiotemporally coupled quantum fuzzy control module includes a quantum state fuzzification unit, a four-dimensional decision matrix construction unit, and a dynamic weight adjustment unit.

[0046] The quantum state fuzzification unit is used to implement fuzzification processing of the quantum state superposition model.

[0047] The four-dimensional decision matrix construction unit is used to build a four-dimensional decision matrix.

[0048] The dynamic weight adjustment unit is used to adjust the decision weights of each dimension based on the Lyapunov index.

[0049] The holographic physical field sensing and driving control module includes a flow field feature extraction unit, a field effect control law generation unit, and an anomaly blocking unit.

[0050] The flow field feature extraction unit is used to extract flow field vortex features.

[0051] The field effect control law generation unit is used to generate the field effect control law.

[0052] The anomaly blocking unit is used to execute the anomaly propagation blocking algorithm;

[0053] The meta-evolutionary fuzzy rule module includes a data fusion unit, a rule competition unit, and a structure adjustment unit.

[0054] The data fusion unit is used to achieve multi-source sensor data fusion in the environmental perception layer.

[0055] The rule competition unit is used to implement rule competition elimination in the evolution-driven layer.

[0056] The structural adjustment unit is used to implement the rule-based structural adjustment of the structural reconstruction layer.

[0057] Preferably, an electronic control device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the control method as described above.

[0058] Preferably, a computer-readable storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor to implement the control method as described above.

[0059] Compared with existing technologies, the beneficial effects of this invention are as follows: The intelligent energy-saving control method for gas-fired boiler steam-water systems based on fuzzy control solves the shortcomings of existing technologies in terms of data fuzziness, spatiotemporal correlation, and anomaly handling, achieving precise control and energy-saving optimization. Specifically, traditional fuzzy control methods often use a single fuzzy set partitioning when dealing with data fuzziness, which makes it difficult to accurately characterize the fuzzy states of each parameter in a complex system and their interrelationships. This method uses the quantum state superposition model |μ>=α|Low>+β|Medium>+γ|High> to map each parameter in the real-time operating data to the quantum state space, and uses quantum probability amplitudes α, β, and γ to accurately represent the different states of the parameters. The method assesses the likelihood of different states and utilizes quantum entanglement to achieve parallel fuzzification of multiple parameters, generating a quantum fuzzy set. This approach more accurately reflects the true operating state of the system, providing a basis for precise control. Existing technologies typically employ planar fuzzy rules when dealing with spatiotemporal correlations, which struggle to effectively capture the complex dynamic changes of the system in time and space. This method establishes a four-dimensional decision matrix with time and space coordinates as dimensions, replacing traditional planar fuzzy rules. Each matrix element corresponds to a control rule at a specific spatiotemporal point, and its value is determined based on the parameter correlation degree at that point. Simultaneously, based on the Lyapunov exponent, a dynamic weight transfer algorithm is used to adjust the decision weights of each dimension in real time. This further enhances the system's dynamic adaptability to spatiotemporal correlations. This spatiotemporally coupled decision matrix comprehensively considers the operating parameters of the gas-fired boiler's steam-water system at different times and spatial locations, as well as their mutual influences. This achieves global optimization and control of the entire system, avoiding the problem of poor overall system performance caused by local optimization in traditional methods. Existing technologies often suffer from untimely detection and delayed processing when handling anomalies, leading to the rapid spread of anomalies and serious damage to the system. This method, upon detecting local overheating anomalies, can promptly activate the anomaly propagation blocking algorithm. Through backpropagation of the control equations, combined with infrared thermal imaging data, it identifies the location and temperature of the local overheated area. The invention solves the control equations to obtain the optimal value of the blocking coefficient β, and controls the actuator to inject reverse energy flow to suppress overheat propagation. In addition, the deformable rule unit library constructed by the meta-evolutionary fuzzy rule module has a self-organizing mechanism during operation, including an environment perception layer, an evolutionary driving layer, and a structure reconstruction layer. It can dynamically adjust the rule unit library according to the complexity and diversity of abnormal situations to generate optimal abnormal handling rules. This efficient abnormal handling capability can block the propagation of abnormalities in a timely and effective manner, avoid causing greater damage to the system, and ensure the safe and stable operation of the system. This invention realizes precise control and energy-saving optimization of the gas boiler steam-water system, improves the system's operating efficiency and stability, and reduces operating costs. Attached Figure Description

[0060] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other embodiments can be derived from the provided drawings without creative effort.

[0061] Figure 1 This is a diagram illustrating the steps of the method of the present invention;

[0062] Figure 2 This is a structural diagram of the device of the present invention. Detailed Implementation

[0063] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent.

[0064] To better illustrate this embodiment, some parts in the accompanying drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions;

[0065] It will be understood by those skilled in the art that certain well-known structures and their descriptions may be omitted in the accompanying drawings.

[0066] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0067] Example 1

[0068] Please see Figure 1 A method for intelligent energy-saving control of a gas-fired boiler steam-water system based on fuzzy control, the method comprising:

[0069] Acquire historical and real-time operating data sets of the gas-fired boiler steam-water system. The historical operating data set includes combustion efficiency, temperature field distribution, vibration frequency, and energy consumption data at different time periods. The real-time operating data set includes current flow field vortex characteristics, acoustic emission signals, infrared thermal image data, and thermal parameters.

[0070] The real-time running data set is input into the quantum state superposition model |μ>=α|Low>+β|Medium>+γ|High> to perform fuzzification processing on the data, where |μ> represents the fuzzified data state, |Low>, |Medium>, and |High> represent the low, medium, and high fuzzy sets, respectively, and α, β, and γ are quantum probability amplitudes satisfying |α| 2 +|β| 2 +|γ| 2 =1;

[0071] A four-dimensional decision matrix is ​​established as a spatiotemporal correlation controller. The four-dimensional decision matrix uses time and space three-dimensional coordinates as dimensions to replace the traditional planar fuzzy rules.

[0072] Based on the Lyapunov index, the decision weights of each dimension of the four-dimensional decision matrix are adjusted in real time using a dynamic weight transfer algorithm. The weight adjustment formula is as follows:

[0073] ω i =exp(λ i ) / ∑exp(λ j )

[0074] Where λ i Let ω be the Lyapunov exponent of the i-th dimension. i These are the decision weights for the corresponding dimensions;

[0075] Combustion intensity is adaptively adjusted by generating a field-effect control law based on the vortex characteristics of the flow field. The formula for the field-effect control law is as follows:

[0076]

[0077] Where Ω is the vortex intensity. For the temperature gradient, k1, k2, and k3 are proportionality coefficients;

[0078] When a local overheating anomaly is detected, the anomaly propagation blocking algorithm is activated to suppress the local overheating in a timely manner through backpropagation of the control equation, which is:

[0079] Where v is the fluid velocity, α is the thermal diffusivity, β is the blocking coefficient, δ is the Dirac function, and r0 is the location of the anomaly source;

[0080] The meta-evolutionary fuzzy rules are invoked, and the meta-evolutionary fuzzy rules construct a deformable rule unit library, whose runtime self-organizing mechanism includes:

[0081] Environmental perception layer: fuses data from multiple sources of sensors, including vibration sensors, acoustic emission sensors, and infrared sensors;

[0082] Evolution-driven layer: A game-theoretic rule-based competition and elimination mechanism filters rule units;

[0083] Structural Reconstruction Layer: Dynamically adjusts the rule structure through a topological gene recombination algorithm;

[0084] Based on the above processing, a load resource regulation scheme is generated to carry out intelligent energy-saving regulation of the gas boiler steam-water system.

[0085] The quantum state superposition model performs data fuzzification processing including:

[0086] Map each parameter in the real-time running data set to the quantum state space, and determine the quantum probability amplitudes α, β, and γ corresponding to each parameter;

[0087] By leveraging the properties of quantum entanglement, parallel fuzzification of multiple parameters is achieved, generating quantum fuzzy sets.

[0088] The establishment of a four-dimensional decision matrix as a spatiotemporal correlation controller includes:

[0089] The matrix is ​​defined by time t and spatial coordinates x, y, and z.

[0090] Each matrix element corresponds to a control rule for a specific spatiotemporal point, and its value is determined based on the parameter correlation degree of that spatiotemporal point.

[0091] The method of generating a field effect control law based on the vortex characteristics of the flow field to adaptively adjust the combustion intensity includes:

[0092] A flow field model is constructed using computational fluid dynamics, and characteristic parameters of the flow field vortex are extracted.

[0093] Based on the mapping relationship between vortex characteristic parameters and combustion efficiency, the parameters of the field effect control law are dynamically adjusted.

[0094] When a local overheating anomaly is detected, the anomaly propagation blocking algorithm is activated to suppress local overheating in a timely manner through backpropagation of the control equations, including:

[0095] The location and temperature of localized overheated areas can be identified using infrared thermal imaging data.

[0096] Solving the control equation yields the optimal value of the blocking coefficient β, and the control actuator injects a reverse energy flow to suppress overheat propagation.

[0097] The structural reconstruction layer: dynamically adjusts the rule structure through a topological gene recombination algorithm, including:

[0098] The topological structure of regular units is encoded to form regular genes;

[0099] Genetic algorithms are used to perform crossover and mutation operations on regular genes to generate new regular structures.

[0100] The new rule structure is evaluated based on its control effect, and the optimal rule structure is retained.

[0101] An electronic control device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the control method as described above.

[0102] A computer-readable storage medium storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by a processor to implement the control method as described above.

[0103] In a large industrial plant, its gas-fired boiler steam-water system plays a crucial role in providing heat energy for the production process. To achieve energy conservation and consumption reduction, an intelligent energy-saving control method based on fuzzy control is adopted.

[0104] The factory's gas-fired boiler steam-water system is equipped with a sensor network. Regarding historical operating data, combustion efficiency was recorded hourly over the past year, fluctuating between 75% and 85%. Temperature field distribution was monitored by 20 temperature sensors evenly distributed within the furnace, recording temperatures at various points every half hour. Vibration frequency was collected in real-time by vibration sensors installed on the boiler body, recorded once per minute, typically within the 40-90Hz range. Energy consumption data was collected by gas flow meters and electricity meters, with gas and electricity consumption tallied every shift (8 hours). For real-time operating data, the current flow field vortex characteristics were acquired by five vortex sensors installed in the steam-water pipes and furnace. Acoustic emission signals were collected by acoustic emission sensors at a frequency of once per second. Infrared thermal image data was captured in real-time by three infrared thermal imagers installed on the outer wall of the furnace. Thermal parameters such as pressure and flow rate were monitored in real-time by corresponding pressure and flow sensors, and this data was transmitted to the data processing center in real-time.

[0105] The real-time running data set is input into the quantum state superposition model |μ>=α|Low>+β|Medium>+γ|High>. For example, for the vortex intensity parameter of the flow field, it is mapped to the quantum state space through a specific algorithm. The quantum probability amplitudes α=0.3, β=0.4, and γ=0.3 are calculated. Utilizing the quantum entanglement property, multiple parameters such as acoustic emission signal intensity and infrared thermal image temperature deviation are simultaneously subjected to parallel fuzzification processing to quickly generate a quantum fuzzy set, providing a basis for decision-making.

[0106] A four-dimensional decision matrix is ​​established with time t (each time interval is 10 minutes) and spatial coordinates x, y, z (the furnace space is divided into 10×10×10 grid cells) as the four dimensions of the matrix. Each matrix element corresponds to the control rule at a specific spatiotemporal point, and its value is determined based on the correlation between multiple parameters such as combustion efficiency, temperature, and pressure at that spatiotemporal point. For example, at the 20th minute, at the position with spatial coordinates (3, 4, 5), if the combustion efficiency is low and the temperature is high while the pressure is normal, the corresponding matrix element is set to the control rule of increasing ventilation and appropriately reducing the gas supply.

[0107] Based on the Lyapunov index, the decision weights of each dimension of the four-dimensional decision matrix are adjusted in real time using a dynamic weight transfer algorithm, and the Lyapunov index λ in the time dimension is calculated. t =0.1, spatial x-dimensional λ x =0.06, y-dimension λ y =0.07, z-dimension λz =0.08, according to the weighting adjustment formula ω i =exp(λ i ) / ∑exp(λ j The decision weights of each dimension are calculated to ensure that the importance of each dimension in decision-making is reasonably allocated under different operating conditions. For example, during the boiler start-up phase, the time dimension has a relatively large weight, focusing more on the trend of parameter changes over time; while during the stable operation phase, the spatial dimension has an appropriately increased weight, focusing on the parameter differences at different spatial locations.

[0108] A flow field model was constructed using computational fluid dynamics software. Vortex characteristic parameters, such as vortex intensity Ω=6 and vortex rotation direction, were extracted from vortex sensor data. Based on the pre-established mapping relationship between vortex characteristic parameters and combustion efficiency, the field effect control law was determined. The proportionality coefficients k1 = 0.25, k2 = 0.15, and k3 = 0.06 are used to calculate the value of u(t). Based on this value, the opening degree of the gas valve and the speed of the ventilation equipment are automatically adjusted to achieve adaptive adjustment of the combustion intensity and improve the combustion efficiency.

[0109] When infrared thermal imaging data shows localized overheating at a certain location r0 in the furnace, exceeding the set threshold (e.g., exceeding the normal temperature by 20°C), the control equation... Given a fluid velocity v = 2.5 m / s and a thermal diffusivity α = 0.015, the optimal value of the blocking coefficient β is obtained by numerically solving the control equation. Then, the actuators, such as cooling fans and water spray devices installed near the furnace, are controlled to inject reverse energy flow, effectively suppressing overheat propagation and avoiding damage to the boiler equipment.

[0110] Invoking meta-evolutionary fuzzy rules:

[0111] Environmental perception layer: Vibration sensors, acoustic emission sensors and infrared sensors collect data in real time. The data fusion unit fuses the data from these multi-source sensors. For example, when the vibration sensor detects abnormal high-frequency vibration, the acoustic emission sensor captures a signal of a specific frequency, and the infrared thermal imager detects an abnormal increase in temperature at the corresponding location, the data is analyzed to determine possible equipment failures or abnormal operating conditions.

[0112] Evolution-driven layer: Based on a game theory-based rule competition and elimination mechanism, rule units are screened. Under different operating conditions, such as changes in boiler load and fuel quality, the control effects of different rules on indicators such as combustion efficiency, energy consumption, and equipment stability are compared. If a rule does not significantly improve combustion efficiency and increases energy consumption under high load conditions, the probability of using the rule is reduced, and rules with poor effects are gradually eliminated, while rules that can effectively improve system performance are retained and strengthened.

[0113] Structural Reconstruction Layer: The topological structure of the rule unit is encoded to form a rule gene. Based on the genetic algorithm, the rule gene is cross-crossed and mutated to generate a new rule structure. For example, the condition and action parts of the rule for matching ventilation volume and gas supply volume are cross-crossed. The new rule structure is evaluated based on the control effect, such as improved energy saving rate and reduced temperature fluctuation. The optimal rule structure is retained and the rule library is continuously optimized.

[0114] Based on the above series of processes, load resource regulation schemes are generated, such as adjusting the gas supply flow, optimizing the ventilation volume, and adjusting the power of the heating elements. The regulation execution module converts these schemes into specific control commands and sends them to the actuators of the gas boiler steam-water system, such as gas regulating valves, ventilation motors, and heating element controllers, to realize intelligent energy-saving regulation of the gas boiler steam-water system, reduce energy consumption, and improve system operation stability and efficiency.

[0115] Example 2

[0116] Please see Figure 2 A smart energy-saving control device for a gas-fired boiler steam-water system based on fuzzy control, characterized in that the device comprises:

[0117] Data acquisition module: used to acquire historical and real-time operating data sets of the gas-fired boiler steam-water system;

[0118] The spatiotemporally coupled quantum fuzzy control module is used to fuzzify data using a quantum state superposition model, establish a four-dimensional decision matrix, and adjust the decision weights of each dimension based on the Lyapunov exponent using a dynamic weight transfer algorithm.

[0119] Holographic physical field perception and drive control module: used to generate field effect control law based on flow field vortex characteristics, and to start an anomaly propagation blocking algorithm when a local overheating anomaly is detected;

[0120] Meta-evolutionary fuzzy rule module: includes an environment perception layer, an evolutionary driving layer and a structure reconstruction layer, used to build a deformable rule unit library and achieve runtime self-organization;

[0121] Control and execution module: Used to control the steam and water system of the gas boiler according to the generated load resource control scheme.

[0122] The spatiotemporally coupled quantum fuzzy control module includes a quantum state fuzzification unit, a four-dimensional decision matrix construction unit, and a dynamic weight adjustment unit.

[0123] The quantum state fuzzification unit is used to implement fuzzification processing of the quantum state superposition model.

[0124] The four-dimensional decision matrix construction unit is used to build a four-dimensional decision matrix.

[0125] The dynamic weight adjustment unit is used to adjust the decision weights of each dimension based on the Lyapunov index.

[0126] The holographic physical field sensing and driving control module includes a flow field feature extraction unit, a field effect control law generation unit, and an anomaly blocking unit.

[0127] The flow field feature extraction unit is used to extract flow field vortex features.

[0128] The field effect control law generation unit is used to generate the field effect control law.

[0129] The anomaly blocking unit is used to execute the anomaly propagation blocking algorithm;

[0130] The meta-evolutionary fuzzy rule module includes a data fusion unit, a rule competition unit, and a structure adjustment unit.

[0131] The data fusion unit is used to achieve multi-source sensor data fusion in the environmental perception layer.

[0132] The rule competition unit is used to implement rule competition elimination in the evolution-driven layer.

[0133] The structural adjustment unit is used to implement the rule-based structural adjustment of the structural reconstruction layer.

[0134] Similarly, in this large industrial plant, its gas-fired boiler steam-water system is equipped with a complete intelligent energy-saving control device, and the specific operation process is as follows:

[0135] The data acquisition module integrates various sensors and data acquisition devices. The historical operation data acquisition section, through an interface with the factory data center, periodically reads accumulated combustion efficiency data (recorded by an efficiency monitor installed at the chimney outlet), temperature field distribution data (from a uniformly distributed array of temperature sensors within the furnace), vibration frequency data (collected by vibration sensors), and energy consumption data (statistics from gas flow meters and electricity meters) from the database. The real-time operation data acquisition section monitors flow field vortex characteristics in real time using vortex sensors installed in the steam-water pipes and at key locations in the furnace; acoustic emission signals are acquired at high frequency by acoustic emission sensors; infrared thermal image data is continuously captured by an infrared thermal imager on the furnace outer wall; and thermal parameters such as pressure and flow rate are acquired in real time by corresponding high-precision sensors. The acquired data is transmitted in real time to the subsequent processing module via high-speed data transmission lines.

[0136] The spatiotemporally coupled quantum fuzzy control module includes:

[0137] The quantum state fuzzification unit receives real-time operating data from the data acquisition module and uses advanced algorithms to map each parameter, such as the flow field vortex intensity and acoustic emission signal frequency, to the quantum state space. It accurately determines the corresponding quantum probability amplitudes α, β, and γ, and utilizes the quantum entanglement property to achieve multi-parameter parallel and rapid fuzzification processing, generating an accurate quantum fuzzy set, and providing a fuzzy data foundation for decision-making.

[0138] The four-dimensional decision matrix construction unit uses time (e.g., every 12 minutes as a time interval) and spatial coordinates (dividing the furnace space into 8×8×8 grid cells) as dimensions to construct a four-dimensional decision matrix. Through complex correlation analysis algorithms, it determines the control rules corresponding to a specific spatiotemporal point for each matrix element. Its value is determined based on the multi-parameter correlation degree of that spatiotemporal point. For example, at a certain moment and a specific furnace space location, if the temperature is too high and the combustion efficiency shows a downward trend, the corresponding matrix element is set to a control rule that reduces the fuel supply and fine-tunes the ventilation angle.

[0139] The dynamic weight adjustment unit calculates the Lyapunov exponents in each dimension (time, space, and three dimensions) in real time, such as the time dimension λ. t =0.11, spatial x-dimensional λ x =0.07, y-dimension λ y =0.08, z-dimension λ z =0.09, calculate the decision weight of each dimension according to the weight adjustment formula, and dynamically adjust the importance of each dimension of the four-dimensional decision matrix in the decision-making according to different operating conditions. For example, when the boiler load changes suddenly, the weight of the time dimension increases, and more attention is paid to the rapid change of parameters; when the boiler is running normally and stably, the weight of the spatial dimension is appropriately increased to ensure that the parameters at each position in the furnace are uniform and stable.

[0140] The holographic physical field perception and driving control module includes:

[0141] The flow field feature extraction unit uses computational fluid dynamics software, combined with data collected by vortex sensors, to construct a high-precision flow field model and accurately extract vortex feature parameters, including detailed information such as vortex intensity, vortex size, and vortex distribution.

[0142] The field effect control law generation unit generates the field effect control law based on the extracted flow field vortex characteristic parameters and a pre-established precise mapping relationship. The proportional coefficients k1, k2, and k3 are determined through an optimization algorithm to achieve precise adaptive adjustment of combustion intensity.

[0143] When infrared thermal imaging data detects a localized overheated area in the furnace, the anomaly blocking unit quickly applies the control equations... By combining known parameters such as fluid velocity v and thermal diffusivity α, the optimal value of the blocking coefficient β is obtained through an efficient numerical solution algorithm. Then, commands are sent to control actuators, such as cooling spray systems and air cooling devices, to inject reverse energy flow, thereby suppressing overheating propagation in a timely manner and ensuring the safe and stable operation of boiler equipment.

[0144] The meta-evolutionary fuzzy rules module includes:

[0145] The data fusion unit performs deep fusion processing on data from vibration sensors, acoustic emission sensors, and infrared sensors. Utilizing advanced data fusion algorithms, such as neural network fusion algorithms, it correlates and analyzes the characteristics of data from different sensors. For example, it combines abnormal vibration signals with the location and temperature change trends of high-temperature areas displayed in infrared thermography to accurately determine the equipment's operating status and potential faults.

[0146] The rule competition unit is based on a game theory-based rule competition and elimination mechanism. Under different operating conditions, such as different production loads and different ambient temperatures, the rule units are strictly screened. Through comprehensive evaluation of multiple indicators, such as combustion efficiency improvement rate, energy consumption reduction rate, and equipment wear degree, the control effect of different rules is compared. For rules with poor performance, their priority in the rule base is gradually reduced, or even eliminated. For rules with excellent performance, their usage frequency and weight are increased, and the rule composition of the rule base is continuously optimized.

[0147] The structural adjustment unit encodes the topology of the rule unit to form rule genes. An advanced genetic algorithm is used to perform crossover and mutation operations on the rule genes to generate new rule structures. The new rule structures are evaluated based on the control effects, such as system response time and energy utilization rate, and the optimal rule structure is retained. This enables the rule base to continuously evolve and optimize in order to adapt to the ever-changing operating environment and requirements.

[0148] The control execution module receives the load resource control scheme generated by the spatiotemporally coupled quantum fuzzy control module, the holographic physical field perception-driven control module, and the meta-evolutionary fuzzy rule module. Through a high-precision control command conversion algorithm, the control scheme is converted into specific control commands and sent to the actuators of the gas boiler steam-water system, such as gas regulating valves, ventilation motor speed controllers, and heating element power controllers, to accurately execute the control operations. This enables intelligent energy-saving control of the gas boiler steam-water system, effectively reducing energy consumption, improving system operating efficiency and stability, and meeting the high-efficiency and energy-saving requirements of factory production.

[0149] The same or similar labels correspond to the same or similar parts;

[0150] The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent.

[0151] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all implementation methods here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A method for intelligent energy-saving control of a gas-fired boiler steam-water system based on fuzzy control, characterized in that, The method includes: Acquire historical and real-time operating data sets of the gas-fired boiler steam-water system. The historical operating data set includes combustion efficiency, temperature field distribution, vibration frequency, and energy consumption data at different time periods. The real-time operating data set includes current flow field vortex characteristics, acoustic emission signals, infrared thermal image data, and thermal parameters. The real-time running data set is input into the quantum state superposition model |μ>=α|Low>+β|Medium>+γ|High> to perform fuzzification processing on the data, where |μ> represents the fuzzified data state, |Low>, |Medium>, and |High> represent the low, medium, and high fuzzy sets, respectively, and α, β, and γ are quantum probability amplitudes satisfying |α| 2 +|β| 2 +|γ| 2 =1; A four-dimensional decision matrix is ​​established as a spatiotemporal correlation controller. The four-dimensional decision matrix uses time and space three-dimensional coordinates as dimensions to replace the traditional planar fuzzy rules. Based on the Lyapunov index, the decision weights of each dimension of the four-dimensional decision matrix are adjusted in real time using a dynamic weight transfer algorithm. The weight adjustment formula is as follows: oh i =exp(λ i ) / ∑exp(λ j ) Where λ i Let ω be the Lyapunov exponent of the i-th dimension. i These are the decision weights for the corresponding dimensions; Combustion intensity is adaptively adjusted by generating a field-effect control law based on the vortex characteristics of the flow field. The formula for the field-effect control law is as follows: Where Ω is the vortex intensity. For the temperature gradient, k1, k2, and k3 are proportionality coefficients; When a local overheating anomaly is detected, the anomaly propagation blocking algorithm is activated to suppress the local overheating in a timely manner through backpropagation of the control equation, which is: Where v is the fluid velocity, α is the thermal diffusivity, β is the blocking coefficient, δ is the Dirac function, and r0 is the location of the anomaly source; The meta-evolutionary fuzzy rules are invoked, and the meta-evolutionary fuzzy rules construct a deformable rule unit library, whose runtime self-organizing mechanism includes: Environmental perception layer: fuses data from multiple sources of sensors, including vibration sensors, acoustic emission sensors, and infrared sensors; Evolution-driven layer: A game-theoretic rule-based competition and elimination mechanism filters rule units; Structural Reconstruction Layer: Dynamically adjusts the rule structure through a topological gene recombination algorithm; Based on the above processing, a load resource regulation scheme is generated to carry out intelligent energy-saving regulation of the gas boiler steam-water system.

2. The control method according to claim 1, characterized in that, The quantum state superposition model performs data fuzzification processing including: Map each parameter in the real-time running data set to the quantum state space, and determine the quantum probability amplitudes α, β, and γ corresponding to each parameter; By leveraging the properties of quantum entanglement, parallel fuzzification of multiple parameters is achieved, generating quantum fuzzy sets.

3. The control method according to claim 1, characterized in that, The establishment of a four-dimensional decision matrix as a spatiotemporal correlation controller includes: The matrix is ​​defined by time t and spatial coordinates x, y, and z. Each matrix element corresponds to a control rule for a specific spatiotemporal point, and its value is determined based on the parameter correlation degree of that spatiotemporal point.

4. The control method according to claim 1, characterized in that, The method of generating a field effect control law based on the vortex characteristics of the flow field to adaptively adjust the combustion intensity includes: A flow field model is constructed using computational fluid dynamics, and characteristic parameters of the flow field vortex are extracted. Based on the mapping relationship between vortex characteristic parameters and combustion efficiency, the parameters of the field effect control law are dynamically adjusted.

5. The control method according to claim 1, characterized in that, When a local overheating anomaly is detected, the anomaly propagation blocking algorithm is activated to suppress local overheating in a timely manner through backpropagation of the control equations, including: The location and temperature of localized overheated areas can be identified using infrared thermal imaging data. Solving the control equation yields the optimal value of the blocking coefficient β, and the control actuator injects a reverse energy flow to suppress overheat propagation.

6. The control method according to claim 1, characterized in that, The structural reconstruction layer: dynamically adjusts the rule structure through a topological gene recombination algorithm, including: The topological structure of regular units is encoded to form regular genes; Genetic algorithms are used to perform crossover and mutation operations on regular genes to generate new regular structures. The new rule structure is evaluated based on its control effect, and the optimal rule structure is retained.

7. An intelligent energy-saving control device for a gas-fired boiler steam-water system based on fuzzy control, characterized in that, The device includes: Data acquisition module: used to acquire historical and real-time operating data sets of the gas-fired boiler steam-water system; The spatiotemporally coupled quantum fuzzy control module is used to fuzzify data using a quantum state superposition model, establish a four-dimensional decision matrix, and adjust the decision weights of each dimension based on the Lyapunov exponent using a dynamic weight transfer algorithm. Holographic physical field perception and drive control module: used to generate field effect control law based on flow field vortex characteristics, and to start an anomaly propagation blocking algorithm when a local overheating anomaly is detected; Meta-evolutionary fuzzy rule module: includes an environment perception layer, an evolutionary driving layer and a structure reconstruction layer, used to build a deformable rule unit library and achieve runtime self-organization; Control and execution module: Used to control the steam and water system of the gas boiler according to the generated load resource control scheme.

8. The control device according to claim 7, characterized in that, The spatiotemporally coupled quantum fuzzy control module includes a quantum state fuzzification unit, a four-dimensional decision matrix construction unit, and a dynamic weight adjustment unit. The quantum state fuzzification unit is used to implement fuzzification processing of the quantum state superposition model. The four-dimensional decision matrix construction unit is used to build a four-dimensional decision matrix. The dynamic weight adjustment unit is used to adjust the decision weights of each dimension based on the Lyapunov index. The holographic physical field sensing and driving control module includes a flow field feature extraction unit, a field effect control law generation unit, and an anomaly blocking unit. The flow field feature extraction unit is used to extract flow field vortex features. The field effect control law generation unit is used to generate the field effect control law. The anomaly blocking unit is used to execute the anomaly propagation blocking algorithm; The meta-evolutionary fuzzy rule module includes a data fusion unit, a rule competition unit, and a structure adjustment unit. The data fusion unit is used to achieve multi-source sensor data fusion in the environmental perception layer. The rule competition unit is used to implement rule competition elimination in the evolution-driven layer. The structural adjustment unit is used to implement the rule-based structural adjustment of the structural reconstruction layer.

9. An electronic control device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the control method as described in any one of claims 1-7.

Images