A fan energy-saving control method based on energy efficiency optimization

Abstract

This invention discloses a fan energy-saving control method based on energy efficiency optimization, comprising the following steps: collecting fan speed, PWM duty cycle, motor power, equipment temperature, load, airflow demand, and hysteresis to generate an energy efficiency dataset; dividing candidate speed ranges and statistically analyzing power changes, temperature drops, and airflow deviations to form a correlation sequence; constructing individual badger position vectors and generating a population; performing energy efficiency sniffing to calculate fitness and generate search directions; performing position mining and marking inefficient sets; performing inefficient retreat and honey collection to generate target energy-saving control parameters; and performing inertial homing to generate control commands and update the population for the next cycle. This invention enables fans to operate stably with low power consumption, improving energy-saving control accuracy and adaptability to operating conditions.

Description

Technical Field

[0001] This invention relates to the field of fan energy-saving control technology, and in particular to a fan energy-saving control method based on energy efficiency optimization. Background Technology

[0002] Existing fan energy-saving control methods mostly employ temperature threshold speed regulation, increasing fan speed when equipment temperature rises and decreasing fan speed when equipment temperature falls. Some control methods collect data on motor power, equipment load, and airflow demand, adjusting the PWM duty cycle based on changes in power consumption and temperature to reduce energy consumption caused by prolonged high-speed fan operation. Other solutions incorporate optimization algorithms to search for target speed, duty cycle, and runtime, ensuring that fan operation matches cooling requirements.

[0003] Current methods still rely primarily on temperature changes for control, lacking joint statistics on power changes, temperature drops, airflow demand deviations, and temperature response lags within the speed range. This makes it difficult to determine whether an increase in speed leads to a high-power, low-return state. When using common optimization algorithms to search for fan control parameters, candidate solutions falling into inefficient speed ranges are often filtered based on fitness ranking, lacking a proactive retreat operation to move the target speed to a safe, energy-saving range. When similar equipment loads and airflow demands reappear, current methods still require re-searching for control parameters, easily leading to repeated optimization and speed fluctuations, affecting the stability of fan energy-saving control. Therefore, providing an energy-efficiency optimization-based fan energy-saving control method is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] One objective of this invention is to propose a fan energy-saving control method based on energy efficiency optimization. This invention fully utilizes fan energy efficiency data, temperature response hysteresis, and the correlation sequence of speed-power consumption-heat dissipation benefit. Through energy efficiency detection, inefficiency avoidance, optimization by sampling, and inertial homing operation, it achieves adaptive energy-saving control of fan speed, PWM duty cycle, and adjustment range. It has the advantages of low energy consumption, stable speed regulation, strong ability to suppress inefficient speed increase, and high efficiency of control parameter reuse.

[0005] According to an embodiment of the present invention, a fan energy-saving control method based on energy efficiency optimization includes the following steps:

[0006] Collect fan speed, PWM duty cycle, motor power, equipment temperature, load, air volume demand, and temperature response hysteresis to generate an energy efficiency dataset;

[0007] Candidate speed intervals are divided into candidate speed intervals, and the power change, temperature drop and air volume demand deviation of each interval are statistically analyzed. A correlation sequence is formed in ascending order of speed.

[0008] A honey badger population is generated by using the center speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and hysteresis threshold to form the individual position vector of the honey badger.

[0009] The energy efficiency sniffing mode is executed, and the fitness value is calculated with power change, air volume demand deviation and lag as penalty items and temperature drop as benefit item, and the search direction is generated in ascending order.

[0010] Execute the mining mode to update the position vector, and write the intervals where the power change is greater than the power consumption growth threshold, the temperature drop is less than the heat dissipation benefit threshold, and the hysteresis is greater than the hysteresis threshold into the inefficient set.

[0011] Execute the inefficient retreat mode, delete the inefficient set and the interval where the predicted temperature is greater than the safety limit, select the remaining interval with the smallest absolute value of the difference with the target speed as the migration speed interval, correct the target speed to the center speed of the migration speed interval, and execute the honey collection mode to generate the target energy-saving control parameters.

[0012] The system executes the inertial homing mode, generates instructions according to the target energy-saving control parameters, collects the power change, temperature drop and hysteresis after execution and writes them into the historical sample library, and selects the historical sample with the smallest sum of the differences with the current load and air volume demand to write into the population of the next cycle.

[0013] Optionally, the operation of acquiring the temperature response hysteresis includes:

[0014] When the fan drive issues a speed adjustment command, record the start time of adjustment. Continuously read the device temperature and take the sampling time when the device temperature first drops and the drop reaches the preset temperature change threshold as the start time of temperature drop.

[0015] The time interval between the start of temperature drop and the start of adjustment is denoted as the temperature response hysteresis.

[0016] Optionally, the operation of forming the associated sequence includes:

[0017] Read the target fan speed, motor power, equipment temperature, air volume demand, and temperature response hysteresis from the energy efficiency dataset;

[0018] The target fan's allowable speed range is divided into candidate speed intervals by fixed speed intervals.

[0019] Sampling records where the target fan speed falls within the same candidate speed range are grouped into the same range record group;

[0020] Within each interval recording group, the power change, temperature drop, air volume demand deviation, and temperature response lag are calculated and arranged in ascending order according to the center speed of the candidate speed interval to form a speed-power consumption-heat dissipation benefit correlation sequence.

[0021] Optionally, the operation of generating a honey badger population includes:

[0022] Read the candidate speed ranges in the speed-power consumption-heat dissipation benefit correlation sequence, and establish honey badger individuals according to the center speed of the candidate speed range from low to high.

[0023] For each individual honey badger, the center speed of the corresponding candidate speed range is taken as the target speed. The PWM duty cycle that is in the same candidate speed range as the target speed is read from the energy efficiency dataset. The difference between the target speed and the target fan speed at the beginning of the current control cycle is taken as the speed adjustment amplitude.

[0024] Write the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold into the honey badger individual position vector in a fixed field order;

[0025] According to the same candidate speed range number, the individual position vector of the honey badger is bound to the power change, temperature drop, air volume demand deviation and temperature response lag, and the bound individual position vectors of the honey badger are formed into a honey badger population.

[0026] Optionally, the operation of executing the energy efficiency sniffing mode includes:

[0027] Read the location vector of each individual honey badger in the honey badger population, and retrieve the bound power change, temperature drop, air volume demand deviation and temperature response lag according to the candidate speed range number;

[0028] Linear normalization was performed based on the maximum and minimum values ​​of the power change, air volume demand deviation, temperature response lag, and temperature drop within the honey badger population, respectively.

[0029] The penalty term is obtained by adding the normalized power change, air volume demand deviation, and temperature response lag.

[0030] The normalized temperature decrease is taken as the benefit item;

[0031] Subtracting the benefit from the penalty term yields the fitness value for each individual honey badger.

[0032] Arrange honey badger individuals in ascending order of fitness value, and determine the position vector of the honey badger individual at the top of the ranking as the current optimal position;

[0033] Calculate the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference between the current optimal position and the position vectors of honey badger individuals other than the first in the sorting. Combine the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference to form the initial search direction for the corresponding honey badger individual before entering the digging mode.

[0034] Optionally, the operation of executing the mining mode includes:

[0035] Read the initial search direction output by the energy efficiency sniffing mode. The initial search direction consists of the target speed difference, the PWM duty cycle difference, and the speed adjustment amplitude difference.

[0036] Read the target rotation speed, PWM duty cycle, rotation speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and temperature response hysteresis threshold from the individual position vector of the honey badger;

[0037] The target speed, PWM duty cycle, and speed adjustment amplitude are added to the corresponding target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference, respectively. The target speed disturbance value, PWM duty cycle disturbance value, and speed adjustment amplitude disturbance value selected from the preset disturbance range are then superimposed to obtain the updated position vector.

[0038] The target speed in the updated position vector is limited to the range of the target fan's allowed speed, the PWM duty cycle in the updated position vector is limited to the range of the fan drive's allowed duty cycle, and the speed adjustment range in the updated position vector is limited to the range of the allowable adjustment in a single control cycle.

[0039] The restricted target speed is mapped to the candidate speed range, and the power change, temperature drop and temperature response hysteresis are read from the speed-power consumption-heat dissipation benefit correlation sequence according to the candidate speed range number.

[0040] Candidate speed ranges with power changes greater than the power consumption growth threshold, temperature drops less than the heat dissipation benefit threshold, and temperature response hysteresis greater than the temperature response hysteresis threshold are written into the inefficient set.

[0041] Optionally, the operation of executing the inefficient backoff mode includes:

[0042] After the honey badger completes the mining mode, read the individual position vector, extract the target rotation speed from the position vector, and map the target rotation speed to the candidate rotation speed range;

[0043] Determine whether the candidate speed range containing the target speed belongs to an inefficient set;

[0044] When the candidate speed range where the target speed is located belongs to the inefficient set, read the temperature drop corresponding to each candidate speed range in the speed-power consumption-heat dissipation benefit correlation sequence, and subtract the temperature drop from the equipment temperature at the beginning of the current control cycle to obtain the predicted equipment temperature for each candidate speed range.

[0045] Remove candidate speed ranges from the inefficient set from the candidate speed ranges, and remove candidate speed ranges whose predicted equipment temperature is greater than the upper limit of temperature safety, to obtain the remaining candidate speed ranges;

[0046] Calculate the absolute value of the difference between the target speed and the center speed of each remaining candidate speed interval, and determine the remaining candidate speed interval with the smallest absolute value of the difference as the migration speed interval;

[0047] When the absolute value of the difference between two remaining candidate speed ranges is the same, the remaining candidate speed range with the lower center speed is selected as the migration speed range.

[0048] The target speed is rewritten as the center speed of the migration speed range.

[0049] Optionally, the operation of executing the honey-collecting mode includes:

[0050] Based on the center speed of the migration speed range, the PWM duty cycle is rewritten according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is rewritten according to the difference between the center speed of the migration speed range and the target fan speed at the beginning sampling time of the current control cycle, so as to obtain the position vector after retreat.

[0051] Using the migration speed range as the local search range of the honey collection mode, a set of candidate target speeds is generated according to the preset speed step size;

[0052] For each candidate target speed in the candidate target speed set, a matching PWM duty cycle is generated according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is generated according to the difference between the candidate target speed and the target fan speed at the beginning of the current control cycle sampling time.

[0053] The candidate target speed, the matched PWM duty cycle, the speed adjustment range, the power consumption growth threshold, the heat dissipation benefit threshold, and the temperature response hysteresis threshold are combined to form a candidate position vector.

[0054] Calculate the fitness value of each candidate location vector, and select the candidate location vector with the smallest fitness value and the predicted equipment temperature not exceeding the upper limit of the temperature safety limit as the target energy-saving control parameter.

[0055] Optionally, the operation of executing the inertial homing mode includes:

[0056] After executing the fan energy-saving control command at the fan drive end, the motor power, equipment temperature, equipment load, air volume demand and temperature response lag at the end of the execution cycle are collected, and the power change and temperature drop during the execution cycle are calculated.

[0057] The target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold in the target energy-saving control parameters are written into the historical sample library along with the power change, temperature drop, temperature response hysteresis, equipment load, and air volume demand after execution.

[0058] At the start of the next control cycle, the equipment load and air volume demand at the beginning sampling time of the next control cycle are read. The absolute value of the difference between the equipment load and the equipment load in the historical sample library, and the absolute value of the difference between the air volume demand and the air volume demand in the historical sample library are calculated respectively. The two absolute values ​​of difference are added together to obtain the sample matching value.

[0059] The historical samples are arranged in ascending order of their matching values. The first historical sample is selected, and the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold from the first historical sample are written into the honey badger individual position vector in the honey badger population for the next control cycle.

[0060] Optionally, the operation of generating and executing fan energy-saving control commands includes:

[0061] Read the target speed, PWM duty cycle and speed adjustment range from the target energy-saving control parameters, and read the target fan speed at the start sampling time of the current control cycle;

[0062] When the target speed is greater than the target fan speed at the start of the current control cycle sampling time, a speed-up control command carrying the PWM duty cycle is generated.

[0063] When the target speed is less than the target fan speed at the start of the current control cycle sampling time, a speed reduction control command carrying the PWM duty cycle is generated.

[0064] When the target speed is consistent with the target fan speed at the start of the current control cycle sampling time, a hold control command carrying the PWM duty cycle is generated.

[0065] One of the speed-up control command, speed-down control command, and hold control command is selected as the fan energy-saving control command and sent to the fan drive end;

[0066] The fan driver operates the target fan according to the PWM duty cycle, and adjusts the actual speed change of the target fan within the current control cycle to be no greater than the speed adjustment range.

[0067] The beneficial effects of this invention are:

[0068] (1) This invention collects fan speed, PWM duty cycle, motor power, equipment temperature, equipment load, air volume demand and temperature response hysteresis to form a speed-power consumption-heat dissipation benefit correlation sequence, so that fan control no longer relies solely on temperature threshold and can identify the matching state between power consumption increase and heat dissipation benefit in different speed ranges.

[0069] (2) The present invention adds an energy efficiency sniffing mode and an inefficiency avoidance mode to the honey badger algorithm, performs migration processing on the target speed that falls into the inefficiency set, so that the honey badger individual avoids the high power consumption and low benefit speed range, and performs honey collection mode in the migration speed range, thereby improving the search stability and effectiveness of energy-saving control parameters.

[0070] (3) The present invention stores the power change, temperature drop, temperature response lag and target energy-saving control parameters after execution into the historical sample library through the inertial homing mode. The historical samples are reused under similar equipment load and air volume requirements, reducing repeated optimization and speed fluctuation, so that the fan operation has a lower energy consumption and more stable adjustment process. Attached Figure Description

[0071] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0072] Figure 1 This is an overall flowchart of a fan energy-saving control method based on energy efficiency optimization proposed in this invention;

[0073] Figure 2 This is a flowchart illustrating the collaborative processing of inefficient retreat mode and honey collection mode in a fan energy-saving control method based on energy efficiency optimization proposed in this invention.

[0074] Figure 3 This invention presents an inertial homing mode and a flowchart of the execution of fan energy-saving control commands for a fan energy-saving control method based on energy efficiency optimization. Detailed Implementation

[0075] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0076] refer to Figures 1-3 A fan energy-saving control method based on energy efficiency optimization includes the following steps:

[0077] Collect fan speed, PWM duty cycle, motor power, equipment temperature, load, air volume demand, and temperature response hysteresis to generate an energy efficiency dataset;

[0078] Candidate speed intervals are divided into candidate speed intervals, and the power change, temperature drop and air volume demand deviation of each interval are statistically analyzed. A correlation sequence is formed in ascending order of speed.

[0079] A honey badger population is generated by using the center speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and hysteresis threshold to form the individual position vector of the honey badger.

[0080] The energy efficiency sniffing mode is executed, and the fitness value is calculated with power change, air volume demand deviation and lag as penalty items and temperature drop as benefit item, and the search direction is generated in ascending order.

[0081] Execute the mining mode to update the position vector, and write the intervals where the power change is greater than the power consumption growth threshold, the temperature drop is less than the heat dissipation benefit threshold, and the hysteresis is greater than the hysteresis threshold into the inefficient set.

[0082] Execute the inefficient retreat mode, delete the inefficient set and the interval where the predicted temperature is greater than the safety limit, select the remaining interval with the smallest absolute value of the difference with the target speed as the migration speed interval, correct the target speed to the center speed of the migration speed interval, and execute the honey collection mode to generate the target energy-saving control parameters.

[0083] The system executes the inertial homing mode, generates instructions according to the target energy-saving control parameters, collects the power change, temperature drop and hysteresis after execution and writes them into the historical sample library, and selects the historical sample with the smallest sum of the differences with the current load and air volume demand to write into the population of the next cycle.

[0084] In this method, the energy efficiency dataset is established according to the control cycle, and the data in each control cycle corresponds to the same fan energy-saving control process. Fan speed is used to characterize the current mechanical operating state of the target fan, PWM duty cycle is used to characterize the output control intensity of the fan driver, motor power is used to characterize the fan energy consumption level, equipment temperature is used to characterize the thermal state of the heat dissipation object, equipment load is used to characterize the heat generation intensity of the heat dissipation object, air volume demand is used to characterize the air volume that the target fan needs to meet in the current control cycle, and temperature response hysteresis is used to characterize the time required for the equipment temperature to effectively decrease after speed adjustment.

[0085] The candidate speed ranges are divided according to the allowable speed range of the target fan. Each candidate speed range corresponds to a set of power change, temperature drop, and airflow demand deviation. The power change is used to determine the degree of energy consumption increase caused by the speed change, the temperature drop is used to determine the heat dissipation benefit caused by the speed change, and the airflow demand deviation is used to determine the degree of matching between the fan output capacity and the airflow demand.

[0086] The individual position vector of the honey badger is used to carry a set of candidate fan control parameters. The target speed, PWM duty cycle, and speed adjustment range are used as execution control parameters, while the power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold are used as inefficiency judgment parameters. The energy efficiency sniffing mode is used to guide the search direction before the mining mode. The mining mode is used to expand the search range of candidate control parameters. The inefficiency avoidance mode is used to make the target speed leave the inefficiency set. The honey-gathering mode is used to perform local search within the migration speed range. The inertial return mode is used to introduce historical samples that meet the energy-saving requirements into the honey badger population in the next control cycle.

[0087] In this embodiment, the operation of acquiring the temperature response hysteresis includes:

[0088] When the fan drive issues a speed adjustment command, record the start time of adjustment. Continuously read the device temperature and take the sampling time when the device temperature first drops and the drop reaches the preset temperature change threshold as the start time of temperature drop.

[0089] The time interval between the start of temperature drop and the start of adjustment is denoted as the temperature response hysteresis.

[0090] When collecting temperature response hysteresis, after the fan drive issues a speed adjustment command, the controller continuously reads the device temperature sampling value and takes the sampling time when the effective temperature drop condition is first met after the adjustment start time as the temperature drop start time. The effective temperature drop condition is determined by the device temperature drop reaching a preset temperature change threshold, which is used to filter sensor noise and small temperature fluctuations.

[0091] The time interval between the start of adjustment and the start of temperature drop is recorded as the temperature response hysteresis. The temperature response hysteresis, along with fan speed, PWM duty cycle, motor power, equipment temperature, equipment load, and airflow demand within the same control cycle, forms a set of control cycle records in the energy efficiency dataset.

[0092] In this embodiment, the operation of forming the associated sequence includes:

[0093] Read the target fan speed, motor power, equipment temperature, air volume demand, and temperature response hysteresis from the energy efficiency dataset;

[0094] The target fan's allowable speed range is divided into candidate speed intervals by fixed speed intervals.

[0095] Sampling records where the target fan speed falls within the same candidate speed range are grouped into the same range record group;

[0096] Within each interval recording group, the power change, temperature drop, air volume demand deviation, and temperature response lag are calculated and arranged in ascending order according to the center speed of the candidate speed interval to form a speed-power consumption-heat dissipation benefit correlation sequence.

[0097] When forming the speed-power consumption-heat dissipation benefit correlation sequence, the controller reads the target fan speed, motor power, equipment temperature, airflow demand, and temperature response hysteresis from the energy efficiency dataset, and divides the target fan speed range into fixed speed intervals from the lower limit to the upper limit. Each candidate speed interval has a unique candidate speed interval number and center speed.

[0098] Sampling records where the target fan speed falls within the same candidate speed range are grouped into the same range record group. The power change within the range record group is obtained from the motor power change within the range record group, the temperature drop is obtained from the equipment temperature change within the range record group, the airflow demand deviation is obtained from the difference between the corresponding airflow demand of the range record group and the air delivery capacity of the candidate speed range, and the temperature response lag is obtained from the temperature response lag within the range record group.

[0099] The correlation sequence of rotational speed, power consumption, and heat dissipation benefits is arranged in ascending order according to the center rotational speed of the candidate rotational speed interval. The arranged correlation sequence provides a data source for subsequent honey badger individual establishment, fitness value calculation, inefficient set generation, and prediction of device temperature calculation.

[0100] In this embodiment, the operations for generating a honey badger population include:

[0101] Read the candidate speed ranges in the speed-power consumption-heat dissipation benefit correlation sequence, and establish honey badger individuals according to the center speed of the candidate speed range from low to high.

[0102] For each individual honey badger, the center speed of the corresponding candidate speed range is taken as the target speed. The PWM duty cycle that is in the same candidate speed range as the target speed is read from the energy efficiency dataset. The difference between the target speed and the target fan speed at the beginning of the current control cycle is taken as the speed adjustment amplitude.

[0103] Write the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold into the honey badger individual position vector in a fixed field order;

[0104] According to the same candidate speed range number, the individual position vector of the honey badger is bound to the power change, temperature drop, air volume demand deviation and temperature response lag, and the bound individual position vectors of the honey badger are formed into a honey badger population.

[0105] When generating the honey badger population, a honey badger individual is established for each candidate speed range. The target speed of the honey badger individual is derived from the center speed of the candidate speed range, the PWM duty cycle is derived from the sampling records in the energy efficiency dataset that are in the same candidate speed range as the target speed, and the speed adjustment range is derived from the difference between the target speed and the target fan speed at the beginning of the current control cycle sampling time.

[0106] The power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold are used as threshold fields in the honey badger's individual position vector. These fields are then used to determine whether the power change, temperature drop, and temperature response hysteresis meet the inefficiency criteria. The honey badger's individual position vector is arranged in the following order: target speed, PWM duty cycle, speed adjustment amplitude, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold.

[0107] Each honey badger's individual location vector is bound to the candidate speed range number along with power change, temperature drop, airflow demand deviation, and temperature response hysteresis. After binding, the energy efficiency sniffing mode can directly read the energy efficiency data corresponding to each honey badger individual.

[0108] In this embodiment, the operation of executing the energy efficiency sniffing mode includes:

[0109] Read the location vector of each individual honey badger in the honey badger population, and retrieve the bound power change, temperature drop, air volume demand deviation and temperature response lag according to the candidate speed range number;

[0110] Linear normalization was performed based on the maximum and minimum values ​​of the power change, air volume demand deviation, temperature response lag, and temperature drop within the honey badger population, respectively.

[0111] The penalty term is obtained by adding the normalized power change, air volume demand deviation, and temperature response lag.

[0112] The normalized temperature decrease is taken as the benefit item;

[0113] Subtracting the benefit from the penalty term yields the fitness value for each individual honey badger.

[0114] Arrange honey badger individuals in ascending order of fitness value, and determine the position vector of the honey badger individual at the top of the ranking as the current optimal position;

[0115] Calculate the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference between the current optimal position and the position vectors of honey badger individuals other than the first in the sorting. Combine the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference to form the initial search direction for the corresponding honey badger individual before entering the digging mode.

[0116] When executing the energy efficiency sniffing mode, the controller reads the location vector of each individual honey badger in the honey badger population and retrieves the bound power change, temperature drop, airflow demand deviation, and temperature response lag based on the candidate speed range number. The power change, airflow demand deviation, and temperature response lag are included in the penalty item, while the temperature drop is included in the benefit item.

[0117] Linear normalization uses the maximum and minimum values ​​within the same field as a benchmark, bringing power change, airflow demand deviation, temperature response lag, and temperature drop to the same numerical scale. The normalized power change, airflow demand deviation, and temperature response lag are added together to form a penalty term, and the normalized temperature drop forms a benefit term. The fitness value is obtained by subtracting the benefit term from the penalty term.

[0118] After sorting the fitness values ​​from smallest to largest, the position vector of the honey badger at the top of the list is determined as the current optimal position. The target rotational speed, PWM duty cycle, and rotational speed adjustment amplitude at the current optimal position are subtracted from the target rotational speed, PWM duty cycle, and rotational speed adjustment amplitude at the position vector of the honey badger to be updated, respectively, yielding the target rotational speed difference, PWM duty cycle difference, and rotational speed adjustment amplitude difference. These three differences form the initial search direction and serve as the basis for updating the position in the mining mode.

[0119] In this embodiment, the operations for executing the mining mode include:

[0120] Read the initial search direction output by the energy efficiency sniffing mode. The initial search direction consists of the target speed difference, the PWM duty cycle difference, and the speed adjustment amplitude difference.

[0121] Read the target rotation speed, PWM duty cycle, rotation speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and temperature response hysteresis threshold from the individual position vector of the honey badger;

[0122] The target speed, PWM duty cycle, and speed adjustment amplitude are added to the corresponding target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference, respectively. The target speed disturbance value, PWM duty cycle disturbance value, and speed adjustment amplitude disturbance value selected from the preset disturbance range are then superimposed to obtain the updated position vector.

[0123] The target speed in the updated position vector is limited to the range of the target fan's allowed speed, the PWM duty cycle in the updated position vector is limited to the range of the fan drive's allowed duty cycle, and the speed adjustment range in the updated position vector is limited to the range of the allowable adjustment in a single control cycle.

[0124] The restricted target speed is mapped to the candidate speed range, and the power change, temperature drop and temperature response hysteresis are read from the speed-power consumption-heat dissipation benefit correlation sequence according to the candidate speed range number.

[0125] Candidate speed ranges with power changes greater than the power consumption growth threshold, temperature drops less than the heat dissipation benefit threshold, and temperature response hysteresis greater than the temperature response hysteresis threshold are written into the inefficient set.

[0126] When executing the excavation mode, the controller reads the initial search direction and the individual honey badger position vector. The target speed, PWM duty cycle, and speed adjustment amplitude are updated along the directions of the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference, respectively, and the target speed disturbance value, PWM duty cycle disturbance value, and speed adjustment amplitude disturbance value selected from the preset disturbance range are superimposed on them.

[0127] After the update, the position vector enters the boundary constraint processing. The target speed is limited to the allowable speed range of the target fan, the PWM duty cycle is limited to the allowable duty cycle range of the fan driver, and the speed adjustment range is limited to the allowable adjustment range of a single control cycle. After the boundary constraints are completed, the updated target speed is mapped to the candidate speed range.

[0128] The controller reads the power change, temperature drop, and temperature response hysteresis from the speed-power consumption-heat dissipation benefit correlation sequence according to the candidate speed range number, and compares them with the power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold, respectively. If the power change is greater than the power consumption growth threshold, the temperature drop is less than the heat dissipation benefit threshold, or the temperature response hysteresis is greater than the temperature response hysteresis threshold, the candidate speed range is written into the inefficient set. The inefficient set stores the candidate speed ranges.

[0129] In this embodiment, the operations for executing the inefficient retreat mode include:

[0130] After the honey badger completes the mining mode, read the individual position vector, extract the target rotation speed from the position vector, and map the target rotation speed to the candidate rotation speed range;

[0131] Determine whether the candidate speed range containing the target speed belongs to an inefficient set;

[0132] When the candidate speed range where the target speed is located belongs to the inefficient set, read the temperature drop corresponding to each candidate speed range in the speed-power consumption-heat dissipation benefit correlation sequence, and subtract the temperature drop from the equipment temperature at the beginning of the current control cycle to obtain the predicted equipment temperature for each candidate speed range.

[0133] Remove candidate speed ranges from the inefficient set from the candidate speed ranges, and remove candidate speed ranges whose predicted equipment temperature is greater than the upper limit of temperature safety, to obtain the remaining candidate speed ranges;

[0134] Calculate the absolute value of the difference between the target speed and the center speed of each remaining candidate speed interval, and determine the remaining candidate speed interval with the smallest absolute value of the difference as the migration speed interval;

[0135] When the absolute value of the difference between two remaining candidate speed ranges is the same, the remaining candidate speed range with the lower center speed is selected as the migration speed range.

[0136] The target speed is rewritten as the center speed of the migration speed range.

[0137] When executing the inefficient retreat mode, the controller reads the individual honey badger's position vector after completing the digging mode and maps the target rotational speed in the position vector to a candidate rotational speed range. If the candidate rotational speed range containing the target rotational speed belongs to the inefficient set, retreat processing is initiated. If the candidate rotational speed range containing the target rotational speed does not belong to the inefficient set, the honey badger enters the honey-gathering mode.

[0138] After the backoff process is initiated, the controller reads the temperature drop corresponding to each candidate speed interval in the speed-power consumption-heat dissipation benefit correlation sequence, and subtracts the temperature drop from the device temperature at the start of the current control cycle to obtain the predicted device temperature for each candidate speed interval. Candidate speed intervals in the inefficient set are removed from the candidate speed interval list, and candidate speed intervals whose predicted device temperature exceeds the temperature safety upper limit are also removed. The remaining candidate speed intervals form the residual candidate speed interval list.

[0139] The controller calculates the absolute value of the difference between the target speed and the center speed of each remaining candidate speed interval, and determines the remaining candidate speed interval with the smallest absolute difference as the migration speed interval. When there are two remaining candidate speed intervals with the same absolute difference, the remaining candidate speed interval with the lower center speed is determined as the migration speed interval. The target speed is rewritten as the center speed of the migration speed interval, and the honey badger individual leaves the candidate speed interval corresponding to the inefficient set.

[0140] In this embodiment, the operation of executing the honey collection mode includes:

[0141] Based on the center speed of the migration speed range, the PWM duty cycle is rewritten according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is rewritten according to the difference between the center speed of the migration speed range and the target fan speed at the beginning sampling time of the current control cycle, so as to obtain the position vector after retreat.

[0142] Using the migration speed range as the local search range of the honey collection mode, a set of candidate target speeds is generated according to the preset speed step size;

[0143] For each candidate target speed in the candidate target speed set, a matching PWM duty cycle is generated according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is generated according to the difference between the candidate target speed and the target fan speed at the beginning of the current control cycle sampling time.

[0144] The candidate target speed, the matched PWM duty cycle, the speed adjustment range, the power consumption growth threshold, the heat dissipation benefit threshold, and the temperature response hysteresis threshold are combined to form a candidate position vector.

[0145] Calculate the fitness value of each candidate location vector, and select the candidate location vector with the smallest fitness value and the predicted equipment temperature not exceeding the upper limit of the temperature safety limit as the target energy-saving control parameter.

[0146] When executing the honey-collecting mode, the controller rewrites the PWM duty cycle and speed adjustment amplitude based on the center speed of the migration speed range. The PWM duty cycle is obtained from a preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is obtained from the difference between the center speed of the migration speed range and the target fan speed at the start of the current control cycle sampling time. The rewritten target speed, PWM duty cycle, speed adjustment amplitude, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold form the position vector after retreat.

[0147] The honey-collecting mode uses the migration speed range as the local search range and generates a set of candidate target speeds according to a preset speed step size. Each candidate target speed in the set is located within the migration speed range. Each candidate target speed generates a matching PWM duty cycle according to a preset speed-PWM duty cycle calibration relationship, and generates a speed adjustment amplitude based on the difference between the candidate target speed and the target fan speed at the start of the current control cycle sampling time.

[0148] Candidate position vectors are composed of candidate target speed, matched PWM duty cycle, speed adjustment amplitude, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold. The controller calculates the fitness value of each candidate position vector and selects the candidate position vector with the smallest fitness value and the predicted equipment temperature not exceeding the upper limit of temperature safety as the target energy-saving control parameter.

[0149] In this embodiment, the operation of executing the inertial homing mode includes:

[0150] After executing the fan energy-saving control command at the fan drive end, the motor power, equipment temperature, equipment load, air volume demand and temperature response lag at the end of the execution cycle are collected, and the power change and temperature drop during the execution cycle are calculated.

[0151] The target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold in the target energy-saving control parameters are written into the historical sample library along with the power change, temperature drop, temperature response hysteresis, equipment load, and air volume demand after execution.

[0152] At the start of the next control cycle, the equipment load and air volume demand at the beginning sampling time of the next control cycle are read. The absolute value of the difference between the equipment load and the equipment load in the historical sample library, and the absolute value of the difference between the air volume demand and the air volume demand in the historical sample library are calculated respectively. The two absolute values ​​of difference are added together to obtain the sample matching value.

[0153] The historical samples are arranged in ascending order of their matching values. The first historical sample is selected, and the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold from the first historical sample are written into the honey badger individual position vector in the honey badger population for the next control cycle.

[0154] When executing the inertial homing mode, after the fan drive completes the fan energy-saving control command, the controller collects the motor power, equipment temperature, equipment load, air volume demand and temperature response lag at the end of the execution cycle, and calculates the power change and temperature drop based on the sampling records at the beginning and end of the execution cycle.

[0155] The target energy-saving control parameters, including target speed, PWM duty cycle, speed adjustment range, power consumption increase threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold, are stored in the historical sample library along with the power change, temperature drop, temperature response hysteresis, equipment load, and airflow demand after execution. Each historical sample in the historical sample library contains control parameter fields and execution result fields.

[0156] At the start of the next control cycle, the controller reads the equipment load and airflow demand at the initial sampling time of the next control cycle, and calculates the absolute value of the difference between the equipment load and the equipment load in the historical sample library, and the absolute value of the difference between the airflow demand and the airflow demand in the historical sample library. The two absolute values ​​of difference are added together to obtain the sample matching value. The historical samples are arranged in ascending order of sample matching value, and the historical sample at the top of the list is selected into the badger population for the next control cycle. The target speed, PWM duty cycle, speed adjustment amplitude, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold of the top-ranked historical sample are written into the badger individual position vector in the badger population for the next control cycle.

[0157] In this embodiment, the operations for generating and executing fan energy-saving control commands include:

[0158] Read the target speed, PWM duty cycle, and speed adjustment range from the target energy-saving control parameters, and read the target fan speed at the start sampling time of the current control cycle;

[0159] When the target speed is greater than the target fan speed at the start of the current control cycle sampling time, a speed-up control command carrying the PWM duty cycle is generated.

[0160] When the target speed is less than the target fan speed at the start of the current control cycle sampling time, a speed reduction control command carrying the PWM duty cycle is generated.

[0161] When the target speed is consistent with the target fan speed at the start of the current control cycle sampling time, a hold control command carrying the PWM duty cycle is generated.

[0162] One of the speed-up control command, speed-down control command, and hold control command is selected as the fan energy-saving control command and sent to the fan drive end;

[0163] The fan driver operates the target fan according to the PWM duty cycle, and adjusts the actual speed change of the target fan within the current control cycle to be no greater than the speed adjustment range.

[0164] When generating and executing fan energy-saving control commands, the controller reads the target speed, PWM duty cycle, and speed adjustment range from the target energy-saving control parameters, and also reads the target fan speed at the start of the current control cycle sampling time. If the target speed is greater than the target fan speed at the start of the current control cycle sampling time, the controller generates an acceleration control command carrying the PWM duty cycle; if the target speed is less than the target fan speed at the start of the current control cycle sampling time, the controller generates a deceleration control command carrying the PWM duty cycle; if the target speed is the same as the target fan speed at the start of the current control cycle sampling time, the controller generates a hold control command carrying the PWM duty cycle.

[0165] The control command that matches the target speed comparison result among the speed-up control command, speed-down control command, and hold control command is determined as the fan energy-saving control command. After the fan energy-saving control command is sent to the fan driver, the fan driver drives the target fan according to the PWM duty cycle and limits the actual speed change of the target fan within the current control cycle according to the speed adjustment range. The actual speed change does not exceed the speed adjustment range.

[0166] Example 1: To verify the feasibility of this invention in practice, it was applied to a heat dissipation scenario of an electronic control cabinet. The control cabinet contains multiple sets of electronic power devices and a target fan. The target fan's speed is adjusted by a PWM duty cycle output from the fan driver. The control cabinet operates under three conditions: low load, medium load, and high load, with the equipment temperature fluctuating according to the load. Traditional temperature threshold speed control directly increases the fan speed after the equipment temperature rises, without distinguishing between the heat dissipation benefit and the increase in power consumption resulting from the increased speed. This easily causes the target fan to remain in the high-speed range for extended periods, resulting in a high-power, low-return operating state.

[0167] In this embodiment, the controller collects fan speed, PWM duty cycle, motor power, equipment temperature, equipment load, air volume demand, and temperature response lag to form an energy efficiency dataset. The target fan's allowable speed range is divided into fixed speed intervals, and the power change, temperature drop, air volume demand deviation, and temperature response lag within each candidate speed interval are statistically analyzed to form a speed-power consumption-heat dissipation benefit correlation sequence.

[0168] The controller establishes a honey badger population based on candidate speed ranges. Each honey badger's position vector consists of the target speed, PWM duty cycle, speed adjustment amplitude, power consumption increase threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold, and is bound to energy efficiency data within the corresponding candidate speed range. In energy efficiency sniffing mode, the controller uses power change, airflow demand deviation, and temperature response hysteresis as penalty terms and temperature decrease as a benefit term to calculate the fitness value and generate an initial search direction. In mining mode, the honey badger's position vector is updated along the initial search direction, and a preset perturbation value is added, enabling the honey badger to search for energy-saving control parameters within the candidate speed range.

[0169] After the mining mode ends, the controller maps the updated target speed to the candidate speed range. Candidate speed ranges with power changes greater than the power consumption growth threshold, temperature drops less than the heat dissipation benefit threshold, and temperature response hysteresis greater than the temperature response hysteresis threshold are written into the inefficient set. When the target speed falls into the inefficient set, the controller executes the inefficiency avoidance mode, deleting the corresponding range in the inefficient set and the range where the predicted device temperature is greater than the upper temperature safety limit. The remaining range with the smallest absolute value of the difference from the target speed is selected as the migration speed range, and the target speed is rewritten as the center speed of the migration speed range.

[0170] After completing the inefficient backoff operation, the controller executes the honeycomb sampling mode, generating a set of candidate target speeds using the migration speed range as the local search area, and matching the PWM duty cycle and speed adjustment amplitude to each candidate target speed. The controller calculates the fitness value of the candidate position vectors and selects the candidate position vector with the smallest fitness value and the predicted equipment temperature not exceeding the upper limit of the temperature safety limit as the target energy-saving control parameter. The fan driver generates control commands based on the target energy-saving control parameters and drives the target fan to run according to the PWM duty cycle.

[0171] After control execution is complete, the controller collects the motor power, equipment temperature, equipment load, airflow demand, and temperature response lag. It calculates the power change and temperature drop within the execution cycle and stores the target energy-saving control parameters, power change, temperature drop, temperature response lag, equipment load, and airflow demand in the historical sample library. At the start of the next control cycle, the controller reads the current equipment load and airflow demand, performs difference matching with the equipment load and airflow demand in the historical sample library, selects the historical sample with the smallest matching value, and writes the target speed, PWM duty cycle, speed adjustment amplitude, power consumption increase threshold, heat dissipation benefit threshold, and temperature response lag threshold from the matching sample into the next control cycle's "honey badger population" (a database of parameters used in the system). This allows the "honey badger population" to start searching from near historically effective control parameters, reducing redundant optimization, ineffective speed regulation, and speed fluctuations under similar operating conditions.

[0172] To verify the control effect, this invention was compared with temperature threshold speed regulation, PID temperature speed regulation, and ordinary honey badger algorithm control under three operating conditions: low load, medium load, and high load. For each operating condition, the average motor power, single-cycle power consumption, maximum equipment temperature, average temperature response lag, airflow demand deviation, proportion of inefficient speeds, number of speed adjustments per cycle, and energy saving rate were statistically analyzed. The results are shown in the table below.

[0173] Table 1: Comparison of Fan Energy-Saving Control Performance

[0174]

[0175] The data in the table shows that under low load conditions, the average motor power of this invention is 14.8W, lower than the 22.4W required for temperature threshold speed regulation. Single-cycle power consumption decreased from 5.60Wh to 3.70Wh, achieving an energy saving rate of 33.9%. The proportion of inefficient speeds decreased from 36.5% to 4.9%, indicating that the inefficiency avoidance mode can identify high-power, low-return speed ranges and shift the target speed to the remaining candidate speed ranges. The number of speed adjustments per cycle decreased from 8.7 to 3.1, indicating that the inertial homing mode reuses historical samples under similar loads and airflow requirements, reducing frequent speed adjustments caused by repeated optimization.

[0176] Under medium load conditions, the highest equipment temperature of this invention is 46.7℃, lower than that of temperature threshold speed regulation, PID temperature speed regulation, and ordinary honey badger algorithm control. The average temperature response lag is 29.5s, lower than the 37.6s of ordinary honey badger algorithm control. These data indicate that the energy efficiency sniffing mode guides the search direction before the mining mode, allowing individual honey badger units to preferentially enter candidate speed ranges with low power change, small airflow demand deviation, and high temperature drop. The airflow demand deviation of this invention is 6.4%, significantly lower than the 19.4% of temperature threshold speed regulation, improving the matching degree between fan output capacity and heat dissipation requirements.

[0177] Under high load conditions, the average motor power of this invention is 31.5W, the single-cycle power consumption is 7.88Wh, and the highest equipment temperature is 54.9℃, which is lower than the three comparative control methods. High load conditions require significant heat dissipation, and ordinary speed reduction strategies can easily lead to temperature safety risks. This invention eliminates candidate speed ranges where the predicted equipment temperature exceeds the upper limit of temperature safety, ensuring that the target energy-saving control parameters meet heat dissipation safety requirements. The proportion of inefficient speeds decreased from 28.3% of temperature threshold speed regulation to 6.5%, indicating that the setting of inefficient sets and migration speed ranges can suppress ineffective speed increases. Considering all three operating conditions, this invention, while maintaining equipment temperature safety, reduces motor power and single-cycle power consumption, reduces airflow demand deviation and speed fluctuations, verifying its feasibility and energy-saving effect.

[0178] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A fan energy-saving control method based on energy efficiency optimization, characterized in that, Includes the following steps: Collect fan speed, PWM duty cycle, motor power, equipment temperature, load, air volume demand, and temperature response hysteresis to generate an energy efficiency dataset; Candidate speed intervals are divided into candidate speed intervals, and the power change, temperature drop and air volume demand deviation of each interval are statistically analyzed. A correlation sequence is formed in ascending order of speed. A honey badger population is generated by using the center speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and hysteresis threshold to form the individual position vector of the honey badger. The energy efficiency sniffing mode is executed, and the fitness value is calculated with power change, air volume demand deviation and lag as penalty items and temperature drop as benefit item, and the search direction is generated in ascending order. Execute the mining mode to update the position vector, and write the intervals where the power change is greater than the power consumption growth threshold, the temperature drop is less than the heat dissipation benefit threshold, and the hysteresis is greater than the hysteresis threshold into the inefficient set. Execute the inefficient retreat mode, delete the inefficient set and the interval where the predicted temperature is greater than the safety limit, select the remaining interval with the smallest absolute value of the difference with the target speed as the migration speed interval, correct the target speed to the center speed of the migration speed interval, and execute the honey collection mode to generate the target energy-saving control parameters. The system executes the inertial homing mode, generates instructions according to the target energy-saving control parameters, collects the power change, temperature drop and hysteresis after execution and writes them into the historical sample library, and selects the historical sample with the smallest sum of the differences with the current load and air volume demand to write into the population of the next cycle.

2. The fan energy-saving control method based on energy efficiency optimization according to claim 1, characterized in that, The operation of acquiring the temperature response hysteresis includes: When the fan drive issues a speed adjustment command, record the start time of adjustment. Continuously read the device temperature and take the sampling time when the device temperature first drops and the drop reaches the preset temperature change threshold as the start time of temperature drop. The time interval between the start of temperature drop and the start of adjustment is denoted as the temperature response hysteresis.

3. The fan energy-saving control method based on energy efficiency optimization according to claim 2, characterized in that, The operation of forming the associated sequence includes: Read the target fan speed, motor power, equipment temperature, air volume demand, and temperature response hysteresis from the energy efficiency dataset; The target fan's allowable speed range is divided into candidate speed intervals by fixed speed intervals. Sampling records where the target fan speed falls within the same candidate speed range are grouped into the same range record group; Within each interval recording group, the power change, temperature drop, air volume demand deviation, and temperature response lag are calculated and arranged in ascending order according to the center speed of the candidate speed interval to form a speed-power consumption-heat dissipation benefit correlation sequence.

4. The fan energy-saving control method based on energy efficiency optimization according to claim 3, characterized in that, The steps for generating a honey badger population include: Read the candidate speed ranges in the speed-power consumption-heat dissipation benefit correlation sequence, and establish honey badger individuals according to the center speed of the candidate speed range from low to high. For each individual honey badger, the center speed of the corresponding candidate speed range is taken as the target speed. The PWM duty cycle that is in the same candidate speed range as the target speed is read from the energy efficiency dataset. The difference between the target speed and the target fan speed at the beginning of the current control cycle is taken as the speed adjustment amplitude. Write the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold into the honey badger individual position vector in a fixed field order; According to the same candidate speed range number, the individual position vector of the honey badger is bound to the power change, temperature drop, air volume demand deviation and temperature response lag, and the bound individual position vectors of the honey badger are formed into a honey badger population.

5. The fan energy-saving control method based on energy efficiency optimization according to claim 4, characterized in that, The operation of executing the energy efficiency sniffing mode includes: Read the location vector of each individual honey badger in the honey badger population, and retrieve the bound power change, temperature drop, air volume demand deviation and temperature response lag according to the candidate speed range number; Linear normalization was performed based on the maximum and minimum values ​​of the power change, air volume demand deviation, temperature response lag, and temperature drop within the honey badger population, respectively. The penalty term is obtained by adding the normalized power change, air volume demand deviation, and temperature response lag. The normalized temperature decrease is taken as the benefit item; Subtracting the benefit from the penalty term yields the fitness value for each individual honey badger. Arrange honey badger individuals in ascending order of fitness value, and determine the position vector of the honey badger individual at the top of the ranking as the current optimal position; Calculate the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference between the current optimal position and the position vectors of honey badger individuals other than the first in the sorting. Combine the target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference to form the initial search direction for the corresponding honey badger individual before entering the digging mode.

6. The fan energy-saving control method based on energy efficiency optimization according to claim 5, characterized in that, The operations for executing the mining mode include: Read the initial search direction output by the energy efficiency sniffing mode. The initial search direction consists of the target speed difference, the PWM duty cycle difference, and the speed adjustment amplitude difference. Read the target rotation speed, PWM duty cycle, rotation speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold and temperature response hysteresis threshold from the individual position vector of the honey badger; The target speed, PWM duty cycle, and speed adjustment amplitude are added to the corresponding target speed difference, PWM duty cycle difference, and speed adjustment amplitude difference, respectively. The target speed disturbance value, PWM duty cycle disturbance value, and speed adjustment amplitude disturbance value selected from the preset disturbance range are then superimposed to obtain the updated position vector. The target speed in the updated position vector is limited to the range of the target fan's allowed speed, the PWM duty cycle in the updated position vector is limited to the range of the fan drive's allowed duty cycle, and the speed adjustment range in the updated position vector is limited to the range of the allowable adjustment in a single control cycle. The restricted target speed is mapped to the candidate speed range, and the power change, temperature drop and temperature response hysteresis are read from the speed-power consumption-heat dissipation benefit correlation sequence according to the candidate speed range number. Candidate speed ranges with power changes greater than the power consumption growth threshold, temperature drops less than the heat dissipation benefit threshold, and temperature response hysteresis greater than the temperature response hysteresis threshold are written into the inefficient set.

7. The fan energy-saving control method based on energy efficiency optimization according to claim 6, characterized in that, The operations for executing the inefficient retreat mode include: After the honey badger completes the mining mode, read the individual position vector, extract the target rotation speed from the position vector, and map the target rotation speed to the candidate rotation speed range; Determine whether the candidate speed range containing the target speed belongs to an inefficient set; When the candidate speed range where the target speed is located belongs to the inefficient set, read the temperature drop corresponding to each candidate speed range in the speed-power consumption-heat dissipation benefit correlation sequence, and subtract the temperature drop from the equipment temperature at the beginning of the current control cycle to obtain the predicted equipment temperature for each candidate speed range. Remove candidate speed ranges from the inefficient set from the candidate speed ranges, and remove candidate speed ranges whose predicted equipment temperature is greater than the upper limit of temperature safety, to obtain the remaining candidate speed ranges; Calculate the absolute value of the difference between the target speed and the center speed of each remaining candidate speed interval, and determine the remaining candidate speed interval with the smallest absolute value of the difference as the migration speed interval; When the absolute value of the difference between two remaining candidate speed ranges is the same, the remaining candidate speed range with the lower center speed is selected as the migration speed range. The target speed is rewritten as the center speed of the migration speed range.

8. The fan energy-saving control method based on energy efficiency optimization according to claim 7, characterized in that, The operation of executing the honey collection mode includes: Based on the center speed of the migration speed range, the PWM duty cycle is rewritten according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is rewritten according to the difference between the center speed of the migration speed range and the target fan speed at the beginning sampling time of the current control cycle, so as to obtain the position vector after retreat. Using the migration speed range as the local search range of the honey collection mode, a set of candidate target speeds is generated according to the preset speed step size; For each candidate target speed in the candidate target speed set, a matching PWM duty cycle is generated according to the preset speed-PWM duty cycle calibration relationship, and the speed adjustment amplitude is generated according to the difference between the candidate target speed and the target fan speed at the beginning of the current control cycle sampling time. The candidate target speed, the matched PWM duty cycle, the speed adjustment range, the power consumption growth threshold, the heat dissipation benefit threshold, and the temperature response hysteresis threshold are combined to form a candidate position vector. Calculate the fitness value of each candidate location vector, and select the candidate location vector with the smallest fitness value and the predicted equipment temperature not exceeding the upper limit of the temperature safety limit as the target energy-saving control parameter.

9. A fan energy-saving control method based on energy efficiency optimization according to claim 8, characterized in that, The operation of executing the inertial homing mode includes: After executing the fan energy-saving control command at the fan drive end, the motor power, equipment temperature, equipment load, air volume demand and temperature response lag at the end of the execution cycle are collected, and the power change and temperature drop during the execution cycle are calculated. The target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold in the target energy-saving control parameters are written into the historical sample library along with the power change, temperature drop, temperature response hysteresis, equipment load, and air volume demand after execution. At the start of the next control cycle, the equipment load and air volume demand at the beginning sampling time of the next control cycle are read. The absolute value of the difference between the equipment load and the equipment load in the historical sample library, and the absolute value of the difference between the air volume demand and the air volume demand in the historical sample library are calculated respectively. The two absolute values ​​of difference are added together to obtain the sample matching value. The historical samples are arranged in ascending order of their matching values. The first historical sample is selected, and the target speed, PWM duty cycle, speed adjustment range, power consumption growth threshold, heat dissipation benefit threshold, and temperature response hysteresis threshold from the first historical sample are written into the honey badger individual position vector in the honey badger population for the next control cycle.

10. A fan energy-saving control method based on energy efficiency optimization according to claim 9, characterized in that, The operation of generating and executing fan energy-saving control commands includes: Read the target speed, PWM duty cycle and speed adjustment range from the target energy-saving control parameters, and read the target fan speed at the start sampling time of the current control cycle; When the target speed is greater than the target fan speed at the start of the current control cycle sampling time, a speed-up control command carrying the PWM duty cycle is generated. When the target speed is less than the target fan speed at the start of the current control cycle sampling time, a speed reduction control command carrying the PWM duty cycle is generated. When the target speed is consistent with the target fan speed at the start of the current control cycle sampling time, a hold control command carrying the PWM duty cycle is generated. One of the speed-up control command, speed-down control command, and hold control command is selected as the fan energy-saving control command and sent to the fan drive end; The fan driver operates the target fan according to the PWM duty cycle, and adjusts the actual speed change of the target fan within the current control cycle to be no greater than the speed adjustment range.

Images