Adaptive cooling regulation method for a power plant acoustic enclosure

Abstract

The application belongs to the technical field of real-time control of soundproof cover heat dissipation process, and specifically provides a self-adaptive heat dissipation adjustment method for a soundproof cover of a power plant. The method mainly synchronously collects the temperature inside and outside the cover, the power of the induced draft fan, the opening degree of the sound-absorbing louver, the rotating speed of the exhaust fan, the plant boundary noise and the meteorological parameters; the heat generation inside the cover and the equivalent sound source intensity are inversely calculated in real time, the temperature and noise combined prediction sequence is recursively generated, and the feasible ventilation amount window is inversely calculated; the minimum energy consumption and equipment loss are taken as the target, the optimal control sequence is solved in a rolling manner, the first adjustment increment is output to drive the execution, and the ventilation optimization under the double constraints is realized. On the basis of keeping the double bottom lines of equipment temperature safety and plant boundary noise compliance, the system operation energy consumption and equipment mechanical loss can be simultaneously reduced, so that the economy and stability of the soundproof cover under full working conditions are overall improved.

Description

Technical Field

[0001] This application belongs to the field of real-time control technology of heat dissipation process of soundproof enclosure, and specifically relates to an adaptive heat dissipation adjustment method for soundproof enclosures in power plants. Background Technology

[0002] In the ventilation and heat dissipation control of soundproof enclosures for power plant equipment, a widely used standard procedure is to collect ambient temperature data inside the enclosure by using temperature sensors installed inside the enclosure, and then adjust the opening range of the sound-absorbing louvers and the operating speed of the exhaust fan according to preset temperature thresholds and closed-loop regulation logic to change the ventilation volume inside the enclosure, thereby achieving temperature control inside the enclosure and ensuring the heat dissipation needs of the internal power generation equipment.

[0003] However, power plant soundproof enclosure systems generally possess the inherent physical property of coupling ventilation paths and sound propagation paths. The silencing channels used for airflow are also the propagation paths for noise radiated from internal equipment. Therefore, sound insulation and noise reduction performance and ventilation / heat dissipation capacity are naturally inversely related. This means that increasing ventilation volume inevitably leads to an increase in the sound radiation area of ​​the silencing channels and an increase in airflow-induced noise, resulting in a decrease in the overall sound insulation attenuation of the enclosure. Simultaneously, the leakage of internal equipment noise to the plant boundary increases, and the gain from ventilation and heat dissipation directly reduces the noise reduction effect of the enclosure. Under these conditions, a control mode that unilaterally adjusts ventilation parameters based on heat dissipation needs without constraining sound radiation levels will directly cause the plant boundary noise to exceed emission standards under high equipment load conditions. This is especially problematic during sensitive periods with strict noise control, such as at night, easily leading to environmental penalties for exceeding standards and even forcing generator units to operate at reduced loads, affecting the stability and economy of power generation.

[0004] Therefore, how to break through the traditional technical framework of single temperature control, couple the dynamic process of thermal equilibrium with the outdoor sound propagation attenuation law in multiple physical domains, construct a joint prediction method for indoor temperature and factory boundary noise, and then solve the optimal ventilation regulation strategy that takes into account both operating energy consumption and equipment loss within the feasible domain formed by the dual hard constraints of temperature safety upper limit and factory boundary noise limit, so as to achieve deep synergistic control of heat dissipation and noise reduction, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] This application provides an adaptive heat dissipation adjustment method for soundproof enclosures in power plants, which effectively solves the problem that it is difficult to balance heat dissipation capacity and sound insulation performance in the ventilation adjustment process of existing technologies. Under high load conditions and noise-sensitive periods, noise at the plant boundary is prone to exceed the standard, making it difficult to ensure both equipment operation safety and environmental compliance requirements. While maintaining the dual bottom line of equipment temperature safety and plant boundary noise compliance, it can simultaneously reduce system operating energy consumption and equipment mechanical wear, thereby improving the overall economy and stability of the soundproof enclosure under all operating conditions.

[0006] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application provides an adaptive heat dissipation adjustment method for a power plant soundproof enclosure, comprising: The system can acquire real-time data on the internal temperature of the soundproof enclosure, the active power of the induced draft fan, the opening degree of the sound-absorbing louvers, the speed of the exhaust fan, the noise level at the factory boundary, the ambient wind speed, the ambient wind direction, and the ambient temperature outside the soundproof enclosure.

[0007] Based on the obtained parameters, the heat generation is estimated using the thermal balance relationship, and the equivalent sound source intensity is calculated by using the inverse relationship of sound propagation attenuation.

[0008] Using heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequence and silencing louver opening sequence as control inputs, a joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the factory boundary is generated by recursively calculating using thermal balance relationship and noise outdoor propagation attenuation relationship.

[0009] By comparing the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper limit of temperature and the noise level at the plant boundary with the plant boundary noise limit, the upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment are calculated to generate a feasible ventilation volume window sequence.

[0010] With the goal of minimizing the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain, and with the feasible ventilation volume window sequence as the control constraint and the internal temperature of the soundproof enclosure not exceeding the preset upper temperature limit and the noise level at the plant boundary not exceeding the plant boundary noise limit as the state constraint in the joint prediction sequence, the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence are generated by solving.

[0011] The first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence is used as the adjustment signal output to control the exhaust fan speed and the silencer louver opening.

[0012] Secondly, this application provides an adaptive heat dissipation regulation system for a power plant soundproof enclosure, comprising:

[0013] Parameter acquisition module: used to acquire in real time the internal temperature of the soundproof enclosure, the active power of the exhaust fan, the opening degree of the sound-absorbing louvers, the speed of the exhaust fan, the noise level at the factory boundary, the ambient wind speed, the ambient wind direction, and the ambient temperature outside the soundproof enclosure.

[0014] Thermoacoustic estimation module: It is used to estimate the heat generation based on the acquired parameters using the thermal balance relationship, and to calculate the equivalent sound source intensity using the inverse relationship of sound propagation attenuation.

[0015] Temperature and noise prediction module: It uses heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequence and silencing louver opening sequence as control inputs. It recursively calculates using thermal balance relationship and noise outdoor propagation attenuation relationship to generate a joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the factory boundary.

[0016] Ventilation zone solution module: It is used to compare the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper limit of temperature, and the noise level at the plant boundary with the plant boundary noise limit, and to calculate the upper limit of maximum ventilation volume and the lower limit of minimum ventilation volume at each moment to generate a feasible ventilation volume window sequence.

[0017] The optimal control solution module is used to minimize the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain. It uses the feasible ventilation volume window sequence as the control constraint and the internal temperature of the soundproof enclosure in the joint prediction sequence not exceeding the preset temperature upper limit and the noise level at the plant boundary not exceeding the plant boundary noise limit as the state constraints. It solves and generates the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence.

[0018] Execution output module: Used to output the first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence as an adjustment signal to control the exhaust fan speed and the silencer louver opening.

[0019] Thirdly, this application provides a readable storage medium, comprising: computer program instructions stored in the readable storage medium, wherein the computer program instructions are read and executed by a processor to perform the steps of an adaptive heat dissipation adjustment method for a power plant soundproof enclosure.

[0020] The beneficial effects of this application are: This application adopts an adaptive heat dissipation adjustment scheme that combines sound and heat, integrating equipment heat dissipation management and plant boundary noise constraints into a unified closed-loop control system. The dynamically linked adjustment logic adapts to different operating conditions, promoting the coordinated matching and synchronous response of the two core objectives of heat dissipation and noise reduction. This effectively solves the problem that existing ventilation adjustment technologies struggle to balance heat dissipation capacity and sound insulation performance, easily leading to excessive plant boundary noise during high-load conditions and noise-sensitive periods, and making it difficult to simultaneously ensure equipment operational safety and environmental compliance. While maintaining both equipment temperature safety and plant boundary noise compliance, this application simultaneously reduces system operating energy consumption and equipment mechanical wear, thereby improving the overall economy and stability of the soundproof enclosure under all operating conditions.

[0021] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application may be realized and generated by the structures pointed out in the description and the accompanying drawings. Attached Figure Description

[0022] Figure 1 A schematic flowchart of the adaptive heat dissipation adjustment method for a power plant soundproof enclosure according to this application is shown. Figure 2 A schematic diagram of the process for generating a joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the plant boundary is shown in this application. Figure 3 A schematic diagram of the process for generating a feasible ventilation window sequence in this application is shown; Figure 4 The flowchart illustrating the process of generating the optimal exhaust fan speed sequence and the silencer louver opening sequence in this application is shown. Figure 5 A schematic diagram of the system architecture and hardware layout of this application is shown. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments generated by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Soundproof enclosures for power plants must simultaneously ensure the heat dissipation needs of internal equipment and meet noise emission standards at the plant boundary. However, ventilation channels and sound propagation paths are naturally coupled, and increasing ventilation volume directly reduces the noise reduction capability of the enclosure. Currently, some control schemes rely solely on the internal temperature of the enclosure for adjustment, failing to incorporate plant boundary noise into real-time constraints. During periods of high equipment load operation or strict nighttime noise control, this can easily lead to excessive noise levels at the plant boundary, even forcing units to operate at reduced loads, making it difficult to balance equipment safety, environmental compliance, and operational economy.

[0025] To resolve the trade-off between heat dissipation and noise reduction, this application proposes an adaptive adjustment scheme that integrates the thermal balance mechanism and the sound propagation attenuation law. This method simultaneously collects multi-dimensional operating parameters of heat, sound, and meteorology, and inverts in real time the two core disturbances: heat generation inside the enclosure and equivalent sound source intensity. A joint prediction sequence for internal temperature and boundary noise is generated through recursive calculations, and feasible ventilation volume windows at each time point are calculated based on dual hard constraints. With the goal of minimizing the frequency conversion loss of the exhaust fan and the wear of the louver actuator, the optimal ventilation control sequence is solved in a rolling manner and output for execution. Under the premise of strictly meeting temperature safety and noise limit requirements, the overall operating cost of the system is optimized.

[0026] The implementation methods of this application will be described in detail and objectively below with specific steps.

[0027] In some embodiments, such as Figure 1As shown, this application provides an adaptive heat dissipation adjustment method for a power plant soundproof enclosure, comprising: S1. Real-time acquisition of the internal temperature of the soundproof enclosure, the active power of the induced draft fan, the opening degree of the sound-absorbing louvers, the speed of the exhaust fan, the noise level at the factory boundary, the ambient wind speed, the ambient wind direction, and the ambient temperature outside the soundproof enclosure.

[0028] The internal temperature of the soundproof enclosure is measured by a Class A platinum resistance sensor installed at the motor temperature measurement point inside the enclosure, avoiding the area of ​​direct heat radiation; the opening value of the sound-absorbing louvers ranges from 0 to 1, where 0 represents fully closed and 1 represents fully open; the noise level at the factory boundary is measured by a Class 1 integrating sound level meter installed at the noise measurement point at the factory boundary, and the measured value is the A-weighted equivalent continuous sound level; the ambient wind speed and direction are measured by a mechanical anemometer; the external ambient temperature of the soundproof enclosure is measured by a Class A platinum resistance sensor installed on the air intake side outside the enclosure, avoiding direct sunlight and the influence of exhaust air.

[0029] To cover the sampling frequency requirements of thermal and acoustic processes, all data sampling periods range from 0.5 to 2 seconds, matching the thermal response time constant of the Class A platinum resistance sensor and the integral characteristics of the Class 1 integrating sound level meter. This also adapts to the rate of change of the large inertia thermal process of the soundproof enclosure, avoiding oversampling and increasing the computational burden on the controller. The specific values ​​can be adjusted within this range based on the sensor's response speed. After acquisition, pulse interference is removed by low-pass filtering. The low-pass filter parameters are designed for the strong electromagnetic interference environment of power plants, specifically filtering out peak measurement errors caused by frequency converter radiation, ensuring the input accuracy of subsequent thermal balance and sound attenuation algorithms, and effectively filtering out peak measurement errors caused by electromagnetic interference. The installation positions of the noise measurement points at the plant boundary correspond one-to-one with the spatial orientation of the sound-absorbing louvers. The fixed azimuth angle of the sound propagation path is the measured layout parameter during hardware installation, providing a definite physical boundary for subsequent sound propagation attenuation calculations. By simultaneously acquiring acoustic and meteorological parameters, complete sensing input is provided to simultaneously meet the dual constraints of temperature and noise.

[0030] S2. Based on the obtained parameters, estimate the heat generation using the thermal balance relationship, and calculate the equivalent sound source intensity using the inverse relationship of sound propagation attenuation.

[0031] Heat generation refers to the total heat power generated by the exhaust fan inside the soundproof enclosure per unit time, measured in kW. It is a non-measurable disturbance that characterizes the intensity of the heat load inside the enclosure. Equivalent sound source intensity refers to the A-weighted sound power level of a point source at the location of the sound-absorbing louvers, representing all noise radiated outward through the sound-absorbing louvers inside the soundproof enclosure. It is measured in dB(A) and is a non-measurable disturbance that characterizes the intensity of sound radiation inside the enclosure. By inverting the disturbances in the two physical domains in real time, accurate input is provided for subsequent state prediction.

[0032] S3. Using the heat generation and equivalent sound source intensity as disturbance inputs, and the future exhaust fan speed sequence and the sound-absorbing louver opening sequence as control inputs, the joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the factory boundary is generated by recursively calculating using the thermal balance relationship and the outdoor noise propagation attenuation relationship.

[0033] The future exhaust fan speed sequence and the future silencer louver opening sequence refer to the time sequence of the control quantities to be optimized arranged in the prediction time domain with a fixed control step size, starting from the current control moment. The time constant of the temperature dynamic inside the soundproof enclosure is 3~5 min, the noise dynamic response time is less than 1 s, and the control step size can be 5~15 s. This matches the mechanical response time of the silencer louver electric actuator with the adjustment accuracy of the exhaust fan frequency converter, avoiding frequent operation that could cause wear on the actuator gears. The prediction time domain range can be 8~15 min, matching the thermal time constant of the soundproof enclosure at 3~5 min, covering 2~3 times the thermal dynamic adjustment cycle, ensuring the integrity of the temperature prediction. This prediction time domain range can cover the complete thermal dynamic adjustment process. The control step size matches the actuator response speed to avoid frequent operation. The algorithm time domain parameters correspond one-to-one with the hardware dynamic characteristics and can be adjusted within the range according to the actuator response characteristics and the system thermal inertia.

[0034] The joint prediction sequence refers to a time series that corresponds one-to-one with the control step size time series, and each time step contains two physical quantities: the predicted internal temperature value and the predicted plant boundary noise level value, covering the entire prediction time domain.

[0035] Using real-time disturbances and the current system state as initial conditions, the future thermal and acoustic states are calculated step by step, which can assess the impact of control actions on the two types of constraints in advance and avoid triggering excessive noise in principle.

[0036] S4. Compare the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper limit of temperature, and the noise level at the plant boundary with the plant boundary noise limit. Calculate the upper limit of maximum ventilation volume and the lower limit of minimum ventilation volume at each moment to generate a feasible ventilation volume window sequence.

[0037] The preset upper limit of temperature can be determined based on the insulation class of the exhaust fan inside the soundproof enclosure and the maximum allowable operating ambient temperature specified in the manufacturer's technical documents. This is to ensure that the equipment is not damaged due to overheating and can be adjusted within the range according to the insulation class of the equipment inside the enclosure. The noise limit at the plant boundary can be determined based on the emission standards for the acoustic environment functional zone approved by the corresponding ecological and environmental department of the project.

[0038] The feasible ventilation window refers to the range of ventilation values ​​that can be taken at each prediction time while simultaneously meeting the requirements of temperature not exceeding the limit and noise not exceeding the standard. The lower bound is the minimum ventilation volume that meets the heat dissipation requirements, and the upper bound is the maximum ventilation volume that does not cause noise to exceed the standard. The set of feasible ventilation windows formed by arranging all prediction time in chronological order can be used to obtain the feasible ventilation window sequence, which is consistent with the prediction time domain length.

[0039] Based on the deterministic physical mapping relationship between ventilation volume and heat dissipation, and between ventilation volume and noise radiation, the dual constraints in the state domain are transformed into the value boundaries of the control domain, which can significantly reduce the complexity of subsequent optimization solutions and ensure that all feasible solutions naturally satisfy the dual hard constraints.

[0040] S5. With the goal of minimizing the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain, and with the feasible ventilation volume window sequence as the control constraint, and the internal temperature of the soundproof enclosure not exceeding the preset temperature upper limit and the noise level at the plant boundary not exceeding the plant boundary noise limit in the joint prediction sequence as the state constraint, the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence are generated by solving.

[0041] The frequency conversion loss of the exhaust fan refers to the electrical energy loss generated when the exhaust fan motor is running at a frequency conversion speed, including the copper loss, iron loss and frequency conversion device loss of the motor; the frequency of operation of the silencer louver refers to the cumulative amplitude of the opening adjustment of the silencer louver within the prediction time domain, which is used to characterize the mechanical wear of the louver actuator.

[0042] The weights are determined based on the actual annual operating cost composition of the project. Specifically, the sum of the annual power consumption cost of the exhaust fan and the annual maintenance and depreciation cost of the louver actuator is first calculated to obtain the total annual operating cost of the system. The weight of the frequency conversion loss of the exhaust fan corresponds to the proportion of the annual power consumption cost of the exhaust fan to the total annual operating cost of the system, and the weight of the frequency of the sound-absorbing louver operation corresponds to the proportion of the annual maintenance and depreciation cost of the louver actuator to the total annual operating cost of the system.

[0043] By adopting the mature rolling time-domain quadratic programming solution method in the field of industrial control, the economically optimal strategy can be solved within the feasible region, which can reduce operating costs while meeting safety and environmental protection constraints.

[0044] S6. The first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence is used as the adjustment signal output to control the exhaust fan speed and the silencer louver opening.

[0045] The first control increment refers to the difference between the first control quantity corresponding to the current moment in the optimal control sequence and the current actual operating control quantity. Only the first increment is output in each control cycle, and subsequent control quantities are recalculated using the latest measurement data in the next cycle. The rolling output of the first control quantity can continuously compensate for model errors and sudden disturbances through feedback correction, resulting in stronger anti-interference capabilities.

[0046] Temperature control and noise control share the same set of ventilation control variables, eliminating the need to set independent adjustment logic. It can automatically match the most suitable ventilation conditions in each control cycle, which not only eliminates the risk of equipment overheating due to insufficient heat dissipation, but also avoids noise leakage problems caused by excessive ventilation, while reducing the overall system energy consumption and equipment wear level.

[0047] In some embodiments, estimating the heat generation using a heat balance relationship includes: S2-a1. Calculate the power loss of the induced draft fan to generate its heat dissipation power; obtain the ambient temperature outside the soundproof enclosure, compare the internal temperature of the enclosure with the external ambient temperature to generate the temperature difference between the inside and outside of the enclosure, and convert the heat power of the temperature difference to generate the heat dissipation power of the enclosure structure; normalize the exhaust fan speed and the opening of the silencer louvers, and convert the normalized data for ventilation and heat dissipation based on the temperature difference between the inside and outside of the enclosure to generate the ventilation and heat dissipation power of the enclosure.

[0048] When calculating the heating power of the induced draft fan, most of the electrical energy input during the operation of the induced draft fan is converted into the mechanical energy of flue gas transportation, and the remaining loss is entirely converted into heat dissipated inside the soundproof enclosure. The conversion formula based on the law of conservation of energy is as follows: , This represents the heating power of the induced draft fan. This represents the real-time active power of the induced draft fan, in kW. The coefficient of performance, representing the power loss and heat generation of the induced draft fan, is dimensionless. , Represents wind turbine efficiency. The values ​​representing motor efficiency are all publicly available nameplate parameters from the induced draft fan manufacturers. This formula takes real-time active power and fixed equipment efficiency parameters as input, and outputs the heat power dissipated by the induced draft fan into the enclosure per unit time, reflecting the intensity of the core heat source inside the enclosure.

[0049] When calculating the heat dissipation power of the enclosure structure, first calculate the temperature difference between the inside and outside of the enclosure. The heat dissipation through the structure of the enclosure follows Fourier's law of thermal conduction. The amount of heat conducted is directly proportional to the area of ​​the enclosure wall and the temperature difference between the inside and outside. Therefore, the formula for the heat dissipation power of the enclosure structure is: , This represents the heat dissipation power of the enclosure structure, measured in kW. This represents the conventional heat transfer coefficient of the corresponding metal damping structure soundproof enclosure, in units of... The selection can be based on the actual material structure of the enclosure wall. Represents the total outer surface area of ​​the soundproof enclosure, in units of The formula takes real-time temperature difference and fixed structural parameters as input and outputs the heat power naturally conducted to the outside through the cover wall per unit time, reflecting the natural heat dissipation capacity without relying on ventilation.

[0050] When calculating the ventilation and heat dissipation power of the enclosure, the exhaust fan speed and the opening of the silencer louvers are first normalized. According to the fan similarity law, the exhaust fan's ventilation volume is directly proportional to its speed; therefore, the normalized speed value is... , The normalized rotational speed is represented by 'n', where 'n' represents the measured exhaust fan speed. This represents the rated speed of the exhaust fan; the ventilation flow area of ​​the silencer louvers is proportional to their opening degree, therefore the normalized value of the opening degree is directly taken from the measured opening degree. The value ranges from 0 to 1, representing the ratio of the actual flow area to the maximum flow area. When the louvers are fully closed, the flow area is 0, and there is no ventilation regardless of the fan speed; when the fan stops, the driving force is 0, and there is no forced ventilation regardless of the louver opening degree. Therefore, the actual relative ventilation volume is... This represents the ratio of the actual ventilation volume to the rated maximum ventilation volume.

[0051] Ventilation and heat dissipation follow the energy equation for open systems. The heat removed by ventilation is equal to the air mass flow rate multiplied by the specific heat of air at constant pressure multiplied by the inlet and outlet temperature difference. The conversion formula is: , This represents the ventilation and heat dissipation power of the enclosure, measured in kW. The specific heat at constant pressure of air at room temperature under standard atmospheric pressure is taken as a generally accepted physical property constant. , The density of air at room temperature under standard atmospheric pressure is represented by a generally accepted physical property constant. , This represents the rated speed of the exhaust fan and the rated ventilation volume when the louvers are fully open, in units of... Divide by 3600 as the time unit from h to s. The inputs to this formula are real-time rotational speed, opening degree, temperature difference, and fixed equipment and physical property parameters. The output is the heat power carried out of the hood by ventilation per unit time, which is used to reflect the heat dissipation capacity of forced ventilation.

[0052] S2-a2. Based on the thermal balance superposition relationship, the heat dissipation power of the cover structure and the heat dissipation power of the cover ventilation are merged to generate the external heat dissipation power of the cover.

[0053] The heat dissipation of the enclosure structure and the ventilation heat dissipation are two parallel heat dissipation paths. The total heat dissipation is the sum of the heat dissipation of the two paths, which gives the total heat dissipation of the enclosure to the outside per unit time under the current state.

[0054] S2-a3. Compare the heating power of the induced draft fan with the heat dissipation power of the enclosure to the outside to generate heat.

[0055] Based on the dynamic relationship of thermal balance, and combined with the rate of temperature change inside the hood to correct the estimation error of the induced draft fan's heating power, the final heat output is the total heating power of the induced draft fan, as shown in the following formula: , This represents the heat output, specifically the total heat generation power of the induced draft fan per unit time. The power of the enclosure to dissipate heat to the outside is represented by kW, and C represents the total heat capacity of the air inside the enclosure, in units of kW. The calculation formula is: V represents the effective internal volume of the soundproof enclosure, in units of... ; This refers to the air density at room temperature and pressure. This refers to the specific heat of air at constant volume under normal temperature and pressure. Represents the real-time rate of change of temperature inside the enclosure, in units of The value can be obtained by linearly fitting the temperature measurement values ​​within a recent time window. The window length is determined based on the balance between the smoothness and real-time performance of the rate of change calculation, and can be adjusted within a range according to the temperature fluctuation range on site. The rate of change of internal energy of the air inside the enclosure is equal to the heat source's heating power minus the total external heat dissipation power. Therefore, the accurate heat source heating power can be inferred from the measured heat dissipation and temperature change rate. Essentially, it uses the actual thermal state feedback of the hardware to correct the heat source disturbance parameters, realizing the dynamic correction of the algorithm input by the hardware state and eliminating the estimation error of the fixed loss coefficient.

[0056] In some embodiments, the equivalent sound source intensity is calculated using the inverse relationship of sound propagation attenuation, including:

[0057] S2-b1. Perform vector decomposition on the ambient wind speed and direction to extract the downwind wind speed component along the direction of sound propagation.

[0058] The sound propagation direction is the fixed azimuth angle from the installation position of the silencing louvers to the noise measurement point at the factory boundary. The wind speed is a vector, and the measured wind direction is the direction from which the wind is coming. It is first converted into the azimuth angle of the direction from which the wind is going. , The azimuth angle of the wind, directly measured by an anemometer, is then used to project the wind speed onto the direction of sound propagation through vector dot product calculation, yielding the downwind component. See the formula: , The component of wind speed along the direction of sound propagation, in units of A positive value indicates that the sound propagation attenuation decreases as the wind blows from the soundproof enclosure towards the measuring point, while a negative value indicates that the sound propagation attenuation increases as the wind blows from the measuring point towards the soundproof enclosure. V represents the measured ambient wind speed, and the unit is... , The unit represents the fixed azimuth angle of the soundproof enclosure pointing towards the measuring point. The core meteorological parameters affecting sound propagation attenuation obtained can provide a basis for correcting the calculation of propagation attenuation accurately.

[0059] S2-b2. The sound propagation attenuation is generated by inverting the sound propagation attenuation of the downwind wind speed component and the preset propagation distance.

[0060] The preset propagation distance is the horizontal straight-line distance from the location of the anechoic louver to the measuring point at the factory boundary. The sound propagation attenuation calculation includes two main attenuation terms: geometric divergence attenuation and atmospheric absorption attenuation.

[0061] Geometric divergence attenuation follows the diffusion law of a semi-free field point source: , The value represents the geometric divergence attenuation, in dB(A), where d represents the physical spatial distance from the installation location of the silencing louver to the noise measurement point at the factory boundary, in meters.

[0062] The additional attenuation caused by wind-induced sound refraction is due to the bending of sound rays caused by the near-surface wind speed gradient: when the wind is downwind, the sound rays bend towards the ground, the sound energy concentrates towards the measuring point, the additional attenuation is negative, and the total propagation attenuation decreases; when the wind is upwind, the sound rays bend upwards, the sound energy dissipates into the air, the additional attenuation is positive, and the total propagation attenuation increases. The additional attenuation is linearly negatively correlated with the downwind wind speed component, and the conversion factor ranges from [value missing]. The calibration method was determined by on-site measurement. The calibration process was as follows: under the condition of fixed load of the induced draft fan and stable equivalent sound source intensity, multiple sets of measured data covering downwind, headwind and crosswind were collected simultaneously. The theoretical values ​​of geometric divergence attenuation and atmospheric absorption attenuation were deducted from the measured total propagation attenuation of each set of data to obtain the wind-induced additional attenuation at the corresponding wind speed. Then, the downwind wind speed component was used as the independent variable for linear fitting. The absolute value of the fitting slope is the conversion coefficient of the propagation path.

[0063] Atmospheric absorption attenuation is indirectly affected by the propagation altitude of sound rays: when the wind is downwind, sound rays bend towards the ground, and the air temperature and humidity near the ground along the propagation path are high, resulting in a large absorption coefficient; when the wind is upwind, sound rays bend upwards, and the air at high altitudes along the propagation path is thin, resulting in a small absorption coefficient. A linearly corrected absorption coefficient and atmospheric absorption coefficient can be used. It exhibits a weak linear positive correlation with the downwind speed component, with a base value ranging from 0.002 to 0.004 dB / m, and a wind speed correction factor ranging from [missing value]. The overall absorption coefficient falls within the typical range of 0.001~0.005 dB / m in normal temperature and humidity environments. The appropriate coefficient can be selected within this range based on the site's altitude and average annual temperature and humidity. The atmospheric absorption attenuation is: , This represents the amount of atmospheric absorption and attenuation, measured in dB(A).

[0064] Total sound propagation attenuation This is the sum of geometrical attenuation, wind-induced refraction attenuation, and atmospheric absorption attenuation. By calculating the total attenuation of noise propagating from the outside of the soundproof enclosure to the measuring point at the factory boundary, it can accurately reflect the combined attenuation effect of distance and meteorological conditions on noise.

[0065] S2-b3. The noise level at the factory boundary is superimposed with the sound propagation attenuation to generate the noise level outside the soundproof enclosure; the airflow regeneration noise level at the sound-absorbing louver is calculated based on the current exhaust fan speed; the transmission loss is converted to the sound-absorbing louver opening to generate the sound-absorbing louver insertion loss.

[0066] During sound propagation, the sound level at the receiving point is equal to the sound level at the sound source minus the propagation attenuation. Therefore, the sound level outside the soundproof enclosure can be deduced as follows: , This represents the noise level outside the soundproof enclosure, measured in dB(A). This represents the measured noise level at the plant boundary, expressed in dB(A).

[0067] When airflow passes through the gaps in the silencing louvers, regenerated noise is generated. This noise source is located on the outside of the louvers, and its sound power is proportional to the sixth power of the airflow velocity. Based on the fan similarity law, the flow velocity is positively correlated with the exhaust fan speed. Therefore, the formula for calculating the regenerated noise level is: , The noise level of airflow regeneration at the current rotational speed is represented by dB(A). The noise level of the airflow regeneration of the silencer louvers at rated speed is represented by dB(A), where n is the current exhaust fan speed. This is the rated speed of the exhaust fan.

[0068] The insertion loss of a sound-absorbing louver is the sound level difference between the two sides of the louver, i.e., the sound insulation of the louver, and is directly related to the opening degree: the larger the opening degree, the larger the flow gap, the more sound leakage, and the lower the insertion loss. The insertion loss is the greatest when fully closed, denoted as . The unit is dB(A), and it is at its minimum when fully open, denoted as The unit is dB(A), and the intermediate aperture is calculated using linear interpolation. , Represents openness Insertion loss at that time, in dB(A).

[0069] S2-b4. Subtract the energy contribution of the airflow-regenerated noise level from the noise level outside the soundproof enclosure to obtain the noise component after the sound source inside the enclosure is transmitted; superimpose the transmitted noise component with the insertion loss of the sound-absorbing louvers to generate the equivalent sound source intensity.

[0070] When sound waves pass through the anechoic louvers, the sound level attenuation equals the insertion loss. The total noise on the outside is composed of the transmitted noise inside the enclosure and the noise level inside the enclosure. Therefore, the transmitted noise component is first separated by energy subtraction. ,in, The transmitted noise level, expressed in dB(A), represents the sound level of the sound source inside the enclosure after passing through the anechoic louvers. The equivalent sound source intensity inside the enclosure is then calculated in reverse. This is the sum of the transmitted noise level and the insertion loss at the corresponding aperture: The obtained equivalent sound source intensity is the core sound disturbance input for subsequent sound state calculations, used to reflect the current equipment noise radiation level inside the enclosure.

[0071] In some embodiments, such as Figure 2 As shown, using heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequences and silencing louver opening sequences as control inputs, a joint prediction sequence is generated by recursively calculating using thermal balance relationships and outdoor noise propagation attenuation relationships. This sequence includes the internal temperature of the soundproof enclosure and the noise level at the factory boundary.

[0072] S3-2. For each control step within the preset prediction time domain, iteratively execute the following operations:

[0073] The total number of iterations corresponding to the prediction time domain is the prediction time domain duration divided by the control step duration.

[0074] S3-2-1. Take the predicted internal temperature value of the previous control step, the real-time ambient temperature and heat generation outside the soundproof enclosure, the exhaust fan speed and the opening of the silencer louvers of the current control step, and generate the predicted internal temperature value of the current control step through thermal balance recursive calculation.

[0075] A recursive model is constructed for the closed cavity characteristics of the metal damping composite soundproof enclosure of this application. Based on the law of conservation of energy, the change in internal energy of the air inside the enclosure within a control step is equal to the net heat gain. The recursive formula is as follows: , , These represent the predicted internal temperature values ​​at step k and step (k+1), in degrees Celsius (°C). This represents the real-time estimated heat generation, which is considered constant within the prediction time domain. This represents the structural heat dissipation power at step k. This represents the ventilation and heat dissipation power at step k in the table. The control step size is represented by seconds (s). Multiplying by 1000 converts the kW unit to W. C represents the total heat capacity of the air inside the enclosure. The core parameters of this recursive model, such as heat capacity and heat transfer coefficient, are derived from the structural dimensions and wall panel material properties of the soundproof enclosure. The recursive dual-heat dissipation path structure and forced ventilation perfectly match the hardware heat dissipation architecture of the soundproof enclosure, which combines natural heat dissipation from the closed wall panels with forced ventilation from the sound-absorbing louvers. The model parameters correspond one-to-one with the hardware physical properties. This formula accurately reflects the gradual temperature change characteristics caused by the heat capacity of the soundproof enclosure. By successively calculating the temperature change under each control action, it accurately matches the characteristics of large inertia and slow dynamics in the thermal process.

[0076] S3-2-2. Take the opening of the silencing louver at the current control step, and generate the predicted value of sound insulation after converting the sound insulation amount.

[0077] The insertion loss is calculated based on the louver opening at the current step size to obtain the real-time sound insulation capability of the silencing louvers at each prediction step size. Since the sound propagation speed is 340m / s, the propagation time at the conventional factory boundary propagation distance is much shorter than the control step size. Therefore, the sound process can be regarded as an algebraic relationship without inertia, and no dynamic recursion is required.

[0078] S3-2-3. The predicted values ​​of equivalent sound source intensity and sound insulation are calculated by sound level attenuation to generate the predicted value of transmitted noise. Combined with the airflow regenerated noise sound level corresponding to the current step length exhaust fan speed, the predicted value of the noise sound level outside the soundproof enclosure is generated by sound level energy superposition.

[0079] The linear variation of the insertion loss of the sound-absorbing louver with the opening degree is the acoustic parameter of the laboratory calibration of this sound-absorbing louver. The geometric divergence adopts the semi-free field formula, corresponding to the semi-free field acoustic boundary conditions of the soundproof enclosure installed on the outdoor ground. The transmitted noise level after noise passes through the louver is: ,in This represents the transmitted noise level of the sound source inside the enclosure at step k, in dB(A). Represents the equivalent sound source intensity calculated in real time. This represents the noise reduction louver insertion loss at step k. Combined with the airflow regenerated noise level corresponding to the current step size and rotational speed. The total noise level outside the louvers is obtained by superimposing energy: , This represents the total noise level outside the louver at step k, in dB(A).

[0080] S3-2-4. Take the predicted value of the equivalent sound source intensity outside the soundproof enclosure, the ambient wind speed, the preset propagation distance, and the ambient wind direction, and generate the predicted value of the factory boundary noise level for the current control step size through forward modeling of sound propagation attenuation.

[0081] The wind speed and direction in the predicted time domain are assumed to be constant. The forward modeling of sound propagation attenuation is based on publicly available outdoor sound propagation laws, calculating the total attenuation from the outside of the louvers to the measuring point at the factory boundary, and then extrapolating the sound level at the measuring point based on the sound source level. By predicting the factory boundary noise level under each control action, the impact of the control action on noise indicators can be reflected.

[0082] S3-2-5. Store the predicted internal temperature value and the predicted plant boundary noise level value of the current control step size into the temperature prediction sequence and the plant boundary noise level prediction sequence, respectively.

[0083] S3-2-6. When the number of iterations reaches the number of steps corresponding to the preset prediction time domain, the internal temperature prediction sequence and the factory boundary noise level prediction sequence are spliced ​​together to generate a joint prediction sequence.

[0084] In some embodiments, the predicted value of the equivalent sound source intensity outside the soundproof enclosure, the ambient wind speed, and the preset propagation distance are taken, and a forward modeling calculation of sound propagation attenuation is performed to generate a predicted value of the factory boundary noise level for the current control step, including:

[0085] S3-2-4-1. Perform vector decomposition on the ambient wind speed and direction, and extract the downwind wind speed component along the line connecting the soundproof enclosure to the measuring point at the factory boundary; perform geometric divergence attenuation calculation on the preset propagation distance to generate the geometric divergence attenuation amount.

[0086] The geometric divergence attenuation is calculated using the aforementioned consistent formula for a semi-free field point source. This attenuation term is the most significant source of attenuation for outdoor sound propagation.

[0087] S3-2-4-2. Based on the downwind speed component, calculate the additional attenuation of wind-induced sound refraction; perform atmospheric absorption attenuation calculation on the preset propagation distance and downwind speed component to generate atmospheric absorption attenuation.

[0088] Using the aforementioned calibrated linear conversion relationship, the additional attenuation caused by near-surface wind speed gradient is calculated. The attenuation is negative under downwind conditions and positive under headwind conditions, characterizing the dominant regulatory role of wind in the spatial distribution of sound energy. Using the aforementioned consistent equivalent wind speed correction for the atmospheric absorption coefficient, the atmospheric absorption attenuation is calculated, reflecting the absorption of sound energy by air molecules; this is a secondary correction term for meteorological influences.

[0089] S3-2-4-3. Combine the geometric divergence attenuation, wind-induced sound refraction attenuation, and atmospheric absorption attenuation to generate the total sound propagation attenuation.

[0090] According to the attenuation superposition rule specified in acoustic standards, the total sound propagation attenuation is obtained by directly summing the independent attenuation terms in the logarithmic domain. .

[0091] S3-2-4-4. Calculate the sound level attenuation by combining the predicted value of the equivalent sound source intensity outside the soundproof enclosure with the total sound propagation attenuation, and generate the predicted value of the factory boundary noise level for the current control step.

[0092] During the forward modeling process, the sound level at the receiving point is equal to the sound level at the source minus the total propagation attenuation, according to the formula... Calculate the predicted boundary noise value at the current step size. .

[0093] In some embodiments, such as Figure 3 As shown, the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence is compared with the preset upper temperature limit, and the noise level at the plant boundary is compared with the plant boundary noise limit. The upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment are then calculated to generate a feasible ventilation volume window sequence, including:

[0094] For each time step in the joint prediction sequence, perform the following operations:

[0095] S4-1. Taking the internal temperature prediction value at time k in the joint prediction sequence as not exceeding the preset upper limit of temperature as a constraint, and combining the heat generation at the current time, the heat dissipation power of the cover structure and the temperature difference between the inside and outside of the cover, the minimum ventilation and heat dissipation power required at that time is directly calculated; and the minimum ventilation volume lower limit is generated by converting the minimum ventilation and heat dissipation power into ventilation volume.

[0096] To ensure the predicted internal temperature at time k By substituting the upper temperature limit into the heat balance relationship used in the aforementioned recursive calculation, the minimum ventilation and heat dissipation power required at that moment can be directly calculated. The structural heat dissipation is the unadjustable natural heat dissipation, and the remaining heat dissipation demand is met by ventilation. Therefore, the minimum ventilation cooling power meets the minimum heat dissipation requirement under temperature constraints. Then, the lower limit of the minimum ventilation volume can be derived by using the ventilation cooling formula: , Represents the minimum ventilation volume limit at time k, in units. When the ventilation volume reaches this value, it can provide sufficient heat dissipation to ensure that the temperature does not exceed the limit.

[0097] S4-2. Calculate the deviation between the factory boundary noise limit and the predicted factory boundary noise level at that moment to generate a noise margin value; using the opening of the silencing louvers, the equivalent sound source intensity, and the noise margin value at that moment as constraints, calculate the maximum ventilation volume limit by back-calculating the ventilation volume limit.

[0098] Reference formula for generating noise margin value: , This represents the noise margin value, measured in dB(A). A positive value indicates that the noise level is below the limit with a margin, while a negative value indicates that the noise level exceeds the limit. Since the noise at the plant boundary increases monotonically with the increase of ventilation volume, there is a unique critical ventilation volume. Exceeding this value will result in noise exceeding the limit. This value is the upper limit of the maximum ventilation volume.

[0099] S4-3. Pair the minimum ventilation volume lower limit with the maximum ventilation volume upper limit to form the feasible ventilation volume window at that moment.

[0100] The feasible ventilation volume window at each moment is a closed interval of ventilation volume. When the ventilation volume is taken within this interval, it can simultaneously meet the dual constraints of temperature and noise.

[0101] S4-4. Arrange the feasible ventilation windows for all time periods in chronological order to generate a sequence of feasible ventilation windows, providing control boundaries for subsequent optimization.

[0102] In some embodiments, the maximum ventilation volume upper limit is generated by back-calculating the ventilation volume upper limit using the opening of the anechoic louvers, the equivalent sound source intensity, and the noise margin values ​​at that moment as constraints, including:

[0103] S4-2-1. Obtain the preset initial value of ventilation volume; calculate the noise contribution value at the plant boundary under the ventilation volume based on the initial value of ventilation volume, the opening degree of the sound-absorbing louvers at that moment, and the equivalent sound source intensity.

[0104] The initial ventilation volume is set to 0, corresponding to a completely closed ventilation state, at which point the noise level is lowest. The noise contribution value at the plant boundary under a given ventilation volume is calculated based on the acoustic characteristics of the sound-absorbing louvers and the laws of sound propagation.

[0105] S4-2-2. Compare the noise contribution value at the plant boundary with the noise margin value to determine whether the noise constraint is met: if the constraint is met, increase the ventilation volume; if the constraint is not met, decrease the ventilation volume. After iterative approximation, take the critical ventilation volume that meets the noise constraint as the upper limit of the maximum ventilation volume.

[0106] The iteration adopts the bisection method commonly used in numerical computing, which utilizes the characteristic that noise increases monotonically with ventilation volume to quickly approach the critical value: the iteration step size ranges from 0.5% to 2% of the rated ventilation volume, and the iteration terminates when the deviation between the noise contribution value at the plant boundary and the noise limit is less than 0.1 to 0.2 dB(A). The balance between calculation accuracy and the computing power of the embedded controller can be comprehensively considered. Specifically, it can be adjusted within the range according to the controller hardware performance and control accuracy requirements. The ventilation volume corresponding to this is the upper limit of the maximum ventilation volume.

[0107] Since the same ventilation volume can be achieved through various combinations of rotational speed and opening degree, and the noise levels at the plant boundary differ for different combinations, the worst-case operating condition principle is adopted when calculating the upper limit of the maximum ventilation volume: for a given ventilation volume, all feasible combinations of rotational speed and opening degree are iterated, and the maximum noise level at the plant boundary is taken as the noise level of that ventilation volume, thus ensuring that all feasible combinations meet the noise constraints. When the ventilation volume changes from 0 to the rated value, the noise at the plant boundary under the worst-case operating condition shows a monotonically increasing trend. Therefore, the bisection method can guarantee global convergence. The number of iterations is related to the step size and accuracy requirements, and it can quickly converge to the critical value. The computational load meets the requirements of real-time control.

[0108] In some embodiments, the noise contribution value at the plant boundary under the ventilation volume is calculated from the initial value of the ventilation volume, the opening degree of the silencing louvers at that moment, and the equivalent sound source intensity, including:

[0109] S4-2-1-1. Calculate the sound insulation of the sound-absorbing louvers based on the opening degree of the sound-absorbing louvers at that moment.

[0110] For a given ventilation volume Q, all ventilation volume relationships are satisfied. Rotational speed n and opening degree In different combinations, the proportions of transmitted noise and airflow-regenerated noise levels vary, resulting in differences in the total noise at the plant boundary. To ensure the conservatism of the upper limit of ventilation volume, the worst-case operating condition principle is adopted: traversing the feasible range of rotational speeds, calculating the corresponding louver opening and the plant boundary noise, and taking the maximum value as the plant boundary noise contribution value under that ventilation volume. Under normal operating conditions, the combination of rated speed and minimum opening is the most unfavorable condition; therefore, the rated speed can be directly used to calculate the corresponding opening, quickly obtaining a conservative noise estimate: given ventilation volume Q, the required louver opening at rated speed. The insertion loss is calculated using the same formula as described above. Combined with the airflow regeneration noise level at the rated speed, the most unfavorable noise level at the plant boundary under this ventilation volume can be obtained.

[0111] S4-2-1-2. Calculate the sound level attenuation by combining the equivalent sound source intensity with the sound insulation of the silencing louvers, and generate the noise level outside the soundproof enclosure.

[0112] Using the same sound level attenuation formula as described above, the transmission noise is obtained by subtracting the insertion loss from the equivalent sound source intensity. This is then superimposed with the sound level energy of the airflow regenerated noise at the corresponding rotational speed to obtain the total noise level outside the soundproof enclosure.

[0113] S4-2-1-3. Perform forward modeling calculations on the noise level outside the soundproof enclosure with the preset propagation distance, real-time acquired ambient wind speed, and ambient wind direction to generate the noise contribution value at the factory boundary.

[0114] Using the same forward modeling method as described above, the total propagation attenuation is calculated, and the noise contribution at the plant boundary is obtained by subtracting the total attenuation from the outer sound level.

[0115] In some embodiments, such as Figure 4 and Figure 5 As shown, with the objective of minimizing the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain, and with the feasible ventilation volume window sequence as the control constraint, and the internal temperature of the soundproof enclosure not exceeding the preset upper temperature limit and the noise level at the plant boundary not exceeding the plant boundary noise limit as the state constraints in the joint prediction sequence, the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence are generated by solving the problem, including:

[0116] S5-1. Within the prediction time domain, construct an optimization objective function that minimizes the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers. Convert the upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment in the feasible ventilation volume window sequence into upper and lower limits of the exhaust fan speed and the upper and lower limits of the sound-absorbing louver opening at that moment, respectively, as control constraints. Set the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence as not exceeding the preset upper limit of temperature as the first state constraint, and set the noise level at the plant boundary at each moment as not exceeding the plant boundary noise limit as the second state constraint.

[0117] Specifically, the physical meaning of optimizing the objective function is to predict the total operating cost in the time domain, as shown in the formula: J represents the total operating cost in yuan, and is the minimization objective. This represents the electricity price for industrial use, determined based on the local power grid's real-time electricity price, in yuan. , The depreciation and maintenance cost corresponding to a unit change in the opening degree of the noise-absorbing louver is calculated based on the actuator's design life, procurement costs, and average annual maintenance costs, expressed in yuan per unit opening degree. Represents rotational speed The variable frequency drive (VFD) losses of the exhaust fan are expressed in kW. When the asynchronous motor operates under VFD conditions, the fan shaft power is proportional to the cube of the rotational speed, and the losses are also approximately proportional to the cube of the rotational speed. The formula is: , Represents the rated speed of the exhaust fan The power loss is a parameter provided on the nameplate by the motor manufacturer. It represents the total change in louver opening within the prediction time domain, reflecting the degree of wear and tear on the actuator.

[0118] The optimization objective is constructed based on the physical operating cost data of the exhaust fan and the silencer louvers. Both indicators of the objective function are designed for the operating characteristics of the hardware: the frequency conversion loss term matches the loss characteristics of the asynchronous motor's frequency conversion operation, which is used to reduce the operating energy consumption of the hardware; the louver operation frequency term corresponds to the mechanical wear characteristics of the electric actuator, which is used to extend the service life of the hardware. The control quantity obtained by solving directly corresponds to the adjustment command of the exhaust fan speed and the opening degree of the silencer louvers.

[0119] Control constraints include three categories: first, the ventilation volume at each moment must meet... Second, speed range constraints. The rated speed range of the corresponding exhaust fan frequency converter is the physical limit of the motor hardware; thirdly, the opening range constraint. The corresponding mechanical limit of the full stroke of the silencer louver electric actuator represents the structural boundary of the actuator hardware. The feasible ventilation window is a mathematical transformation of the two core functional constraints of the hardware: the lower bound is derived from the upper temperature limit determined by the insulation class of the motor inside the enclosure, and the upper bound is derived from the sound insulation requirements determined by the noise emission standards at the factory boundary. The algorithm transforms the dual physical constraints of the hardware into feasible boundaries of the control domain, ensuring that all optimal solutions naturally adapt to the performance limits of the hardware. The state constraint is the temperature at each moment. ,noise As a redundant safety constraint, it further ensures reliability. , These represent the preset upper temperature limit and the factory boundary noise limit, respectively.

[0120] S5-2. Under the condition that the control quantity constraint, the first state constraint and the second state constraint are satisfied at the same time, the optimization objective function is solved by rolling time domain optimization to obtain the optimal exhaust fan speed sequence and the silencer louver opening sequence.

[0121] The solution employs a sequential quadratic programming algorithm commonly used in numerical optimization. Since the objective function is convex and the constraints form a convex feasible region, this optimization problem is a convex optimization problem with a unique global optimum and no local extrema. The single-cycle solution time meets the real-time control requirements of embedded systems.

[0122] In some embodiments, the first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence is output as an adjustment signal to control the exhaust fan speed and the silencer louver opening, including:

[0123] S6-1. Extract the first element of the optimal exhaust fan speed sequence, calculate the deviation between this element and the current exhaust fan speed, and generate an incremental exhaust fan speed signal; extract the first element of the optimal silencer louver opening sequence, calculate the deviation between this element and the current silencer louver opening, and generate an incremental silencer louver opening signal.

[0124] The first element of the optimal control sequence is the target control value to be achieved in the current control cycle, and the incremental signal is the difference between the target value and the current actual value. Only the first control variable is executed in each control cycle, and subsequent control variables are re-optimized in the next cycle using the latest measurement data. This is the standard implementation logic for rolling time-domain control, which can continuously correct model errors and disturbance effects.

[0125] S6-2. Convert the exhaust fan speed increment signal to an analog signal and generate a first analog adjustment signal and output it; convert the silencer louver opening increment signal to an analog signal and generate a second analog adjustment signal and output it.

[0126] The output signal is directly connected to the analog input port of the exhaust fan frequency converter and the louver actuator to drive the equipment to adjust to the target state, which is a common signal transmission method for industrial process control.

[0127] In some embodiments, such as Figure 3 As shown, this application provides an adaptive heat dissipation regulation system for a power plant soundproof enclosure, comprising:

[0128] Parameter acquisition module: used to acquire in real time the internal temperature of the soundproof enclosure, the active power of the exhaust fan, the opening degree of the sound-absorbing louvers, the speed of the exhaust fan, the noise level at the factory boundary, the ambient wind speed, the ambient wind direction, and the ambient temperature outside the soundproof enclosure.

[0129] Thermoacoustic estimation module: It is used to estimate the heat generation based on the acquired parameters using the thermal balance relationship, and to calculate the equivalent sound source intensity using the inverse relationship of sound propagation attenuation.

[0130] Temperature and noise prediction module: It uses heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequence and silencing louver opening sequence as control inputs. It recursively calculates using thermal balance relationship and noise outdoor propagation attenuation relationship to generate a joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the factory boundary.

[0131] Ventilation zone solution module: It is used to compare the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper limit of temperature, and the noise level at the plant boundary with the plant boundary noise limit, and to calculate the upper limit of maximum ventilation volume and the lower limit of minimum ventilation volume at each moment to generate a feasible ventilation volume window sequence.

[0132] The optimal control solution module is used to minimize the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain. It uses the feasible ventilation volume window sequence as the control constraint and the internal temperature of the soundproof enclosure in the joint prediction sequence not exceeding the preset temperature upper limit and the noise level at the plant boundary not exceeding the plant boundary noise limit as the state constraints. It solves and generates the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence.

[0133] Execution output module: Used to output the first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence as an adjustment signal to control the exhaust fan speed and the silencer louver opening.

[0134] In some embodiments, this application provides a readable storage medium, comprising: computer program instructions stored in the readable storage medium, wherein the computer program instructions are read and executed by a processor to perform the steps of an adaptive heat dissipation adjustment method for a power plant soundproof enclosure.

[0135] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0136] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0137] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An adaptive heat dissipation adjustment method for a soundproof enclosure in a power plant, characterized in that, include: Real-time data acquisition includes the internal temperature of the soundproof enclosure, the active power of the induced draft fan, the opening degree of the sound-absorbing louvers, the speed of the exhaust fan, the noise level at the factory boundary, the ambient wind speed, the ambient wind direction, and the ambient temperature outside the soundproof enclosure. Based on the obtained parameters, the heat generation is estimated using the thermal balance relationship, and the equivalent sound source intensity is calculated using the inverse relationship of sound propagation attenuation. Using heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequence and silencing louver opening sequence as control inputs, a joint prediction sequence including the internal temperature of the soundproof enclosure and the noise level at the factory boundary is generated by recursively calculating using the thermal balance relationship and the noise outdoor propagation attenuation relationship. By comparing the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper limit of temperature and the noise level at the factory boundary with the factory boundary noise limit, the upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment are calculated to generate a feasible ventilation volume window sequence. With the goal of minimizing the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers in the prediction time domain, and with the feasible ventilation volume window sequence as the control constraint, and the internal temperature of the soundproof enclosure not exceeding the preset temperature upper limit and the noise level at the plant boundary not exceeding the plant boundary noise limit in the joint prediction sequence as the state constraint, the optimal exhaust fan speed sequence and the sound-absorbing louver opening sequence are generated by solving. The first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence is used as the adjustment signal output to control the exhaust fan speed and the silencer louver opening.

2. The method based on claim 1, characterized in that, Estimating heat generation using heat balance relationships includes: The active power of the induced draft fan is converted into power loss to generate the induced draft fan heating power; the ambient temperature outside the soundproof enclosure is obtained, and the internal temperature of the soundproof enclosure is compared with the ambient temperature outside the soundproof enclosure to generate the temperature difference between the inside and outside of the enclosure. The heat power of the temperature difference between the inside and outside of the enclosure is converted to generate the heat dissipation power of the enclosure structure; the exhaust fan speed and the opening of the sound-absorbing louvers are normalized, and the ventilation and heat dissipation conversion is performed on the normalized data based on the temperature difference between the inside and outside of the enclosure to generate the ventilation and heat dissipation power of the enclosure. Based on the thermal balance superposition relationship, the heat dissipation power of the cover structure and the heat dissipation power of the cover ventilation are merged to generate the external heat dissipation power of the cover. The heat generation is calculated by comparing the heat output of the induced draft fan with the heat dissipation power of the enclosure.

3. The method based on claim 1, characterized in that, The equivalent sound source intensity is calculated using the inverse relationship of sound propagation attenuation, including: Vector decomposition of ambient wind speed and direction is performed to extract the downwind wind speed component along the direction of sound propagation; The sound propagation attenuation is generated by inverting the sound propagation attenuation of the downwind wind speed component and the preset propagation distance. The noise level at the factory boundary is superimposed with the sound propagation attenuation to generate the noise level outside the soundproof enclosure; the airflow regeneration noise level at the sound-absorbing louvers is calculated based on the current exhaust fan speed; the transmission loss of the sound-absorbing louvers is converted to generate the insertion loss of the sound-absorbing louvers. The noise level outside the soundproof enclosure is reduced by the energy contribution of the airflow-regenerated noise level to obtain the noise component transmitted through the sound source inside the enclosure. The noise component transmitted through the sound source inside the enclosure is then superimposed with the sound-absorbing louver insertion loss to generate an equivalent sound source intensity.

4. The method based on claim 1, characterized in that, Using heat generation and equivalent sound source intensity as disturbance inputs, and future exhaust fan speed sequences and silencing louver opening sequences as control inputs, a joint prediction sequence is generated by recursively calculating using thermal balance relationships and outdoor noise propagation attenuation relationships. This sequence includes the internal temperature of the soundproof enclosure and the noise level at the plant boundary. Use the current internal temperature of the soundproof enclosure as the initial value for the internal temperature prediction sequence; For each control step within the preset prediction time domain, the following calculation is performed iteratively: The predicted internal temperature of the previous control step is taken, along with the real-time ambient temperature and heat generation outside the soundproof enclosure, the exhaust fan speed and the opening of the silencer louvers at the current control step. Through thermal balance recursion calculation, the predicted internal temperature of the current control step is generated. Take the opening of the silencing louver at the current control step, and generate the predicted sound insulation value after converting the sound insulation amount; The predicted values ​​of equivalent sound source intensity and sound insulation are calculated by sound level attenuation to generate the predicted value of transmitted noise. Combined with the airflow regenerated noise sound level corresponding to the current step length exhaust fan speed, the predicted value of noise sound level outside the soundproof enclosure is generated by sound level energy superposition. The predicted value of the equivalent sound source intensity outside the soundproof enclosure, the ambient wind speed, the preset propagation distance, and the ambient wind direction are taken, and the predicted value of the factory boundary noise level at the current control step is generated by forward modeling the sound propagation attenuation. Store the internal temperature prediction value and the plant boundary noise level prediction value of the current control step size into the temperature prediction sequence and the plant boundary noise level prediction sequence, respectively. Once the number of iterations reaches the preset number of steps corresponding to the prediction time domain, the internal temperature prediction sequence and the factory boundary noise level prediction sequence are spliced ​​together to generate a joint prediction sequence.

5. The method based on claim 4, characterized in that, The predicted equivalent sound source intensity outside the soundproof enclosure, ambient wind speed, and preset propagation distance are used to generate the predicted noise level at the plant boundary for the current control step size through forward modeling of sound propagation attenuation. This includes: Vector decomposition is performed on the ambient wind speed and direction to extract the downwind wind speed component along the line connecting the soundproof enclosure to the measuring point at the factory boundary; geometric divergence attenuation is calculated for the preset propagation distance to generate the geometric divergence attenuation amount. Based on the downwind speed component, the additional attenuation of wind-induced sound refraction is calculated; atmospheric absorption attenuation is calculated for the preset propagation distance and downwind speed component to generate atmospheric absorption attenuation. The geometrical divergence attenuation, wind-induced sound refraction attenuation, and atmospheric absorption attenuation are combined to generate the total sound propagation attenuation. The predicted value of the equivalent sound source intensity outside the soundproof enclosure is used to calculate the sound level attenuation with the total sound propagation attenuation, and the predicted value of the factory boundary noise level for the current control step is generated.

6. The method based on claim 4, characterized in that, By comparing the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence with the preset upper temperature limit and the noise level at the plant boundary with the plant boundary noise limit, the upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment are calculated to generate a feasible ventilation volume window sequence, including: For each time step in the joint prediction sequence, perform the following operations: With the constraint that the predicted internal temperature at time k in the joint prediction sequence does not exceed the preset upper temperature limit, and combined with the heat generation at the current time, the heat dissipation power of the cover structure and the temperature difference between the inside and outside of the cover, the minimum ventilation and heat dissipation power required at that time is directly calculated; and the minimum ventilation and heat dissipation power is converted into ventilation volume to generate the minimum ventilation volume lower limit. The deviation between the factory boundary noise limit and the predicted factory boundary noise level at that moment is calculated to generate the noise margin value; the maximum ventilation volume upper limit is generated by back-calculating the ventilation volume upper limit using the sound-absorbing louver opening, equivalent sound source intensity and noise margin value at that moment as constraints. Pair the minimum ventilation volume lower limit with the maximum ventilation volume upper limit to form the feasible ventilation volume window at that moment; Arrange the feasible ventilation windows at all times in chronological order to generate a sequence of feasible ventilation windows.

7. The method based on claim 6, characterized in that, Using the opening degree of the anechoic louvers, the equivalent sound source intensity, and the noise margin at that moment as constraints, the maximum ventilation volume upper limit is generated by back-calculating the upper limit of ventilation volume, including: Obtain the preset initial value of ventilation volume; calculate the noise contribution value at the plant boundary under the ventilation volume based on the initial value of ventilation volume, the opening degree of the sound-absorbing louvers at that moment, and the equivalent sound source intensity. Compare the noise contribution value at the plant boundary with the noise margin value to determine whether the noise constraint is met: if the constraint is met, increase the ventilation volume; if the constraint is not met, decrease the ventilation volume. After iterative approximation, take the critical ventilation volume that meets the noise constraint as the upper limit of the maximum ventilation volume.

8. The method based on claim 7, characterized in that, The noise contribution at the plant boundary under this ventilation volume is calculated based on the initial ventilation volume, the opening degree of the silencing louvers at that moment, and the equivalent sound source intensity, including: The sound insulation of the sound-absorbing louvers is calculated based on the opening degree of the sound-absorbing louvers at that moment; The sound level attenuation is calculated by combining the equivalent sound source intensity with the sound insulation of the sound-absorbing louvers, and the noise level outside the soundproof enclosure is generated. The noise level outside the soundproof enclosure is used to perform forward modeling calculations of sound propagation attenuation with the preset propagation distance, real-time acquired ambient wind speed, and ambient wind direction to generate the noise contribution value at the factory boundary.

9. The method based on claim 1, characterized in that, With the objective of minimizing the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of operation of the silencer louvers in the prediction time domain, and with the feasible ventilation volume window sequence as the control constraint, and the internal temperature of the soundproof enclosure not exceeding the preset upper temperature limit and the noise level at the plant boundary not exceeding the plant boundary noise limit as the state constraints in the joint prediction sequence, the optimal exhaust fan speed sequence and the silencer louver opening sequence are generated by solving the problem, including: Within the prediction time domain, an optimization objective function is constructed that minimizes the weighted sum of the frequency conversion loss of the exhaust fan and the frequency of the sound-absorbing louvers. The upper limit of the maximum ventilation volume and the lower limit of the minimum ventilation volume at each moment in the feasible ventilation volume window sequence are converted into upper and lower limits of the exhaust fan speed and the opening of the sound-absorbing louvers at that moment, respectively, as control constraints. The first state constraint is that the internal temperature of the soundproof enclosure at each moment in the joint prediction sequence does not exceed the preset upper limit of temperature, and the second state constraint is that the noise level at the plant boundary at each moment does not exceed the plant boundary noise limit. Under the condition that the control quantity constraint, the first state constraint and the second state constraint are satisfied at the same time, the optimization objective function is solved by rolling time domain optimization to obtain the optimal exhaust fan speed sequence and the silencer louver opening sequence.

10. The method based on claim 1, characterized in that, The first control increment of the optimal exhaust fan speed sequence and the silencer louver opening sequence is used as the adjustment signal output to control the exhaust fan speed and the silencer louver opening, including: Extract the first element of the optimal exhaust fan speed sequence, calculate the deviation between this element and the current exhaust fan speed, and generate an incremental exhaust fan speed signal; extract the first element of the optimal silencer louver opening sequence, calculate the deviation between this element and the current silencer louver opening, and generate an incremental silencer louver opening signal. The incremental signal of the exhaust fan speed is converted from digital to analog to generate the first analog adjustment signal and output; the incremental signal of the opening of the silencer louvers is converted from digital to analog to generate the second analog adjustment signal and output.

Images