High-temperature predictive starting fan control system based on energy efficiency optimization

Abstract

The invention provides a high-temperature prediction starting fan control system based on energy efficiency optimization, and relates to the technical field of fan control, firstly, based on a pre-calibrated fan and equipment heat influence weight matrix, the system can accurately evaluate the comprehensive cooling effect under the synergistic effect of multiple fans, the limitation of independent regulation and control of a single fan is overcome, and the efficiency of the system is improved. The utilization efficiency of heat dissipation resources is improved, then the energy efficiency regulation ratio which is a quantitative index is defined, and the overtemperature risk reduction amount and the fan power consumption increment are subjected to normalization comparison, so that the control decision gives consideration to the safety and the energy efficiency, the situation that too high energy consumption cost is paid for tiny risks is avoided, and meanwhile, the energy efficiency is improved. The system activates the evaluation process only when the prediction risk exists, and sets the energy efficiency critical value as the execution threshold, thereby further reducing the unnecessary fan action and calculation overhead, achieving the conversion from passive response to active prediction and from local speed regulation to global optimization, and reducing the energy consumption of the system on the premise of ensuring the reliable operation of the equipment.

Description

Technical Field

[0001] This invention relates to the field of fan control technology, and in particular to a high-temperature predictive start-up fan control system based on energy efficiency optimization. Background Technology

[0002] In data centers, high-performance computing devices, and other systems with high-temperature heat-generating equipment requiring cooling, these devices easily generate significant heat under high loads. Inadequate heat dissipation can cause junction temperatures to exceed safety limits, leading to performance degradation or even permanent damage. Traditional fan control strategies often employ feedback mechanisms based on current temperature; for example, activating a fan or increasing its speed when the temperature exceeds a threshold. These methods suffer from lag, cannot handle temperature rises caused by sudden load changes, and often result in unnecessary energy consumption due to overcooling. Some advanced solutions incorporate temperature prediction, but often only use a fixed safety margin as the control target, failing to quantify the trade-off between over-temperature risk and fan power consumption, making it difficult to achieve optimal energy efficiency. Furthermore, existing systems typically control each fan independently, neglecting the coupled cooling effect of multiple fans on multiple devices, resulting in low control efficiency. There is an urgent need for a fan control system capable of predicting thermal risks in advance, comprehensively assessing the cooling benefits and energy costs of different fan actions, and making intelligent decisions based on global energy efficiency optimization, to significantly reduce heat dissipation power consumption while ensuring equipment safety.

[0003] Therefore, it is necessary to provide a high-temperature predictive start-up fan control system based on energy efficiency optimization to solve the above-mentioned technical problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a high-temperature predictive start-up fan control system based on energy efficiency optimization, achieving the beneficial effect of making intelligent fan control decisions based on global energy efficiency optimization.

[0005] This invention provides a high-temperature predictive start-up fan control system based on energy efficiency optimization, comprising: Load prediction module: Based on the operating load power sequence of each heat-generating device within a preset time period in the past, extrapolate to obtain the predicted operating load power trajectory within a preset time window in the future; Temperature prediction module: Based on a pre-built first-order thermal response model, and using the predicted operating load power trajectory, current ambient temperature and current equipment temperature as inputs, it extrapolates the predicted temperature trajectory within a preset time window in the future; Safety limit calculation module: Based on the predicted operating load power trajectory and the maximum allowable junction temperature limit of the device, calculate the sequence of thermal safety limit for each heat-generating device within a preset time window in the future; Risk Integral Calculation Module: The module compares the predicted temperature trajectory with the thermal safety upper limit sequence point by point, integrates the portion exceeding the limit over time, obtains the over-temperature risk integral value of each heating device, and then sums the over-temperature risk integral values ​​of each heating device by weight to obtain the total over-temperature risk integral value. Energy efficiency assessment module: Based on the pre-built first-order thermal response model and candidate fan control actions, it combines the predicted operating load power trajectory, current ambient temperature and current equipment temperature to deduce the intervention temperature trajectory, and calculates the energy efficiency regulation ratio of each candidate fan control action; Fan control module: Based on the energy efficiency adjustment ratio of each candidate fan control action and the preset energy efficiency threshold, determine the target fan control action and send the gear control command to the corresponding target fan.

[0006] Preferably, in the temperature prediction module, the pre-constructed first-order thermal response model is:

[0007] in, For heating equipment Future No. The device temperature at each point in time. For heating equipment thermal time constant, The current ambient temperature. For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. This is the preset predictive control time step.

[0008] Preferably, in the safety upper limit calculation module, the calculation formula for the sequence of thermal safety upper limits for each heat-generating device within a future preset time window is as follows:

[0009] in, For heating equipment Future No. The thermal safety limit at a given point in time. For heating equipment The device allows for a maximum junction temperature limit. For the temperature rise safety factor, For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time.

[0010] Preferably, in the energy efficiency assessment module, the pre-calibration step of the pre-calibrated fan and equipment thermal influence weight matrix includes: When each heat-generating device is under high load, control all fans to run at the lowest safe speed setting so that all heat-generating devices reach a high temperature steady state. The high temperature steady state is when each heat-generating device is within 80% to 95% of its maximum allowable junction temperature limit, and the temperature change rate does not exceed ±1 degree Celsius per minute for five consecutive minutes. Each fan was individually increased to its preset speed setting, while the remaining fans were kept at their lowest safe speed setting. After the temperature of each heating device has stabilized, record the temperature drop data of each heating device. Based on the maximum temperature drop caused by each fan, the temperature drop data is normalized to obtain the relative cooling weight of each fan for each heat-generating device. The fan and equipment thermal influence weight matrix is ​​obtained by combining all the relative cooling weights into a matrix.

[0011] Preferably, in the energy efficiency assessment module, the formula for predicting the intervention temperature trajectory is as follows:

[0012]

[0013] in, For heating equipment Future No. Predicted intervention temperature at each time point For heating equipment thermal time constant, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. The preset predictive control time step, For all fans in the future At what time point did the heating device Overall cooling effect For the pre-calibrated thermal influence weight matrix of fans and equipment, For pre-calibrated fans The cooling efficiency function of rotational speed, The total number of fans, This is due to the delay in cooling response.

[0014] Preferably, in the energy efficiency assessment module, the calculation steps for the energy efficiency regulation ratio of each candidate fan control action include: The predicted intervention temperature trajectory corresponding to the candidate fan control action is compared point by point with the thermal safety upper limit sequence at each moment in the future time window. The part of the temperature exceeding the thermal safety upper limit is accumulated over time to obtain the intervention over-temperature risk integral value of each heat-generating device. The intervention over-temperature risk integral values ​​of all heat-generating devices are weighted and summed to obtain the total intervention over-temperature risk integral value corresponding to the candidate fan control action. Calculate the difference between the over-temperature risk integral value and the total intervention over-temperature risk integral value corresponding to the candidate fan control action to obtain the over-temperature risk integral reduction achieved by the candidate fan control action; Obtain the total fan power consumption increment caused by the candidate fan control action relative to the current fan operating state; Divide the integral reduction in over-temperature risk achieved by the candidate fan control action by the total fan power consumption increment to obtain the energy efficiency regulation ratio of the candidate fan control action.

[0015] Preferably, the candidate fan control action refers to a control combination consisting of one or more fans at their respective selectable speeds, and the selectable speed of each fan is not lower than the current operating speed.

[0016] Preferably, in the energy efficiency assessment module, the predicted intervention temperature trajectory is only extrapolated for equipment with an over-temperature risk integral value greater than zero.

[0017] Preferably, the energy efficiency assessment module is activated only when the total over-temperature risk integral value is greater than zero. If the total over-temperature risk integral value is not greater than zero, the fan control module maintains the current operating speed of each fan unchanged.

[0018] Preferably, in the fan control module, after determining the target fan control action, if the energy efficiency regulation ratio corresponding to the target fan control action is lower than the preset energy efficiency threshold, then any fan adjustment is abandoned, and the current operating speed of each fan remains unchanged.

[0019] Compared with related technologies, the high-temperature predictive start-up fan control system based on energy efficiency optimization provided by this invention has the following beneficial effects: This invention introduces a mechanism for predicting future loads and temperatures, which can proactively identify potential thermal risks before the heat-generating equipment actually exceeds its temperature, avoiding the lag of traditional feedback control and effectively ensuring the operational safety of key components.

[0020] Firstly, based on the pre-calibrated fan and equipment thermal influence weight matrix, the system can accurately evaluate the comprehensive cooling effect under the coordinated action of multiple fans, overcoming the limitations of independent control of a single fan and improving the utilization efficiency of heat dissipation resources.

[0021] Then, by defining the energy efficiency regulation ratio as a quantitative indicator, the reduction in over-temperature risk is normalized and compared with the increase in fan power consumption, so that the control decision takes into account both safety and energy efficiency, avoiding paying too high an energy cost for minor risks.

[0022] In addition, the system activates the evaluation process only when there is a predicted risk and sets an energy efficiency threshold as an execution threshold, which further reduces unnecessary fan operation and computational overhead. This achieves a shift from passive response to active prediction and from local speed regulation to global optimization, significantly reducing the energy consumption of the cooling system while ensuring reliable equipment operation. Attached Figure Description

[0023] Figure 1 This is a module architecture diagram of the high-temperature predictive start-up fan control system based on energy efficiency optimization of the present invention; Figure 2 This is a schematic diagram of the working process of the high-temperature predictive start-up fan control system based on energy efficiency optimization of the present invention. Detailed Implementation

[0024] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the drawings, not all structures. Moreover, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0025] It should also be noted that, for ease of description, the accompanying drawings show only the parts relevant to the invention and not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it may also have additional steps not included in the drawings. The process may correspond to a method, function, procedure, subroutine, subroutine, etc.

[0026] Example 1 A high-temperature predictive start-up fan control system based on energy efficiency optimization, in its specific implementation, such as... Figure 1 As shown, it illustrates the module architecture diagram of the energy efficiency-optimized high-temperature predictive start-up fan control system of the present invention, including: Load prediction module 100: Based on the operating load power sequence of each heat-generating device within a preset time period in the past, extrapolate to obtain the predicted operating load power trajectory within a preset time window in the future.

[0027] In the specific implementation process, firstly, during the system initialization phase, the control cycle of the entire system is set. The control cycle is set to a fixed value, determined by the heat dissipation performance and thermal inertia characteristics of the equipment, typically 1 second. The operating rhythm of all subsequent modules is based on this control cycle. At the beginning of each control cycle, the load prediction module 100 performs a system initialization phase to configure historical data. For example, the acquisition window length is 10 seconds and the sampling period is 1 second, thus forming a running load power sequence containing 10 consecutive load power values. The load prediction module 100 uses a first-order linear extrapolation algorithm to calculate the predicted trajectory within the next five seconds. For example, the running load power sequence is linearly fitted using the least squares method to obtain an upward trend line with a certain slope, and starting from the last value of the sequence, it is recursively extrapolated second by second according to the determined slope to obtain the predicted running load power trajectory for the next five seconds, providing data basis for subsequent modules.

[0028] Temperature prediction module 200: Based on a pre-built first-order thermal response model, and using the predicted operating load power trajectory, current ambient temperature and current equipment temperature as inputs, it extrapolates the predicted temperature trajectory within a preset time window.

[0029] Specifically, in the temperature prediction module, the pre-built first-order thermal response model is as follows:

[0030] in, For heating equipment Future No. The device temperature at each point in time. For heating equipment thermal time constant, The current ambient temperature. For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. This is the preset predictive control time step.

[0031] In the specific implementation process, the pre-constructed first-order thermal response model obtains key parameters through offline calibration before system deployment. The equipment thermal time constant and equipment thermal resistance are fixed parameters obtained through offline calibration during the system initialization phase. The equipment thermal time constant is obtained by applying a step load to the heat-generating equipment at a constant ambient temperature and recording the temperature response curve, and then fitting an exponential function. The equipment thermal resistance is obtained by measuring the difference between the equipment temperature and the ambient temperature under steady-state high load conditions and then dividing it by the load power. At the beginning of the control cycle, the current equipment temperature and the current ambient temperature are read. At the same time, the load prediction module 100 outputs the predicted operating load power trajectory for the next five seconds. The temperature prediction module 200 substitutes the current equipment temperature, the current ambient temperature, the predicted operating load power trajectory, the thermal time constant, and the equipment thermal resistance into the first-order thermal response model for discrete recursive calculation. First, the predicted temperature at the end of the first second is calculated, and so on, second by second, to obtain the predicted temperature trajectory for the next five seconds, providing a data basis for subsequent steps.

[0032] Safety Upper Limit Calculation Module 300: Based on the predicted operating load power trajectory and the maximum allowable junction temperature limit of the device, calculate the sequence of thermal safety upper limits for each heat-generating device within a preset time window in the future.

[0033] Specifically, in the safety upper limit calculation module, the calculation formula for the thermal safety upper limit sequence of each heat-generating device within a future preset time window is as follows:

[0034] in, For heating equipment Future No. The thermal safety limit at a given point in time. For heating equipment The device allows for a maximum junction temperature limit. For the temperature rise safety factor, For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time.

[0035] In practical implementation, the safety upper limit calculation module 300 is used to calculate the maximum junction temperature based on the predicted operating load power trajectory, the preset maximum junction temperature threshold of the device, the thermal resistance of the equipment, and the temperature rise safety factor. The maximum allowable junction temperature limit of the device is the maximum long-term reliable operating temperature limit specified by each heat-generating equipment manufacturer in its product technical specifications. This maximum allowable junction temperature limit is entered by the user according to the actual equipment model during the system configuration phase. For example, the maximum allowable junction temperature limit for a servo drive device is 125℃; the maximum allowable junction temperature limit for the winding portion monitored by the built-in temperature sensor of the geared motor is 155℃; the maximum allowable junction temperature limit for an industrial motor using Class F insulation is 155℃; the maximum allowable junction temperature limit for an IGBT power module is 150℃; and the maximum allowable junction temperature limit for the mineral oil circuit in a hydraulic system is 65℃. The thermal safety upper limit sequence is generated by calculating the thermal safety upper limit sequence, which corresponds to each moment within a preset time window. The highest junction temperature threshold of the device is the maximum long-term reliable operating temperature limit specified by the equipment manufacturer in the product specifications. The device thermal resistance is the complete thermal resistance parameter from the heat source to the environment obtained through offline calibration. The temperature rise safety factor is a fixed value configured in the system to introduce additional design margin to cope with model uncertainties. The temperature rise safety factor ranges from 0.1 to 0.5, and its specific value is determined through system reliability verification experiments: the equipment is operated under typical load conditions, and the temperature rise safety factor is gradually reduced until an over-temperature protection trigger event occurs. The critical value is recorded, and a 20% safety margin is retained as the final set value. For example, for an industrial servo drive, if the standard deviation of its thermal model prediction error is tested to be 3 degrees Celsius, then the temperature rise safety factor is set to 0.2, making the thermal safety upper limit 20% lower than the theoretical maximum allowable temperature rise, thereby ensuring no misjudgment occurs at a 99.7% confidence level. The safety limit calculation module 300 calculates the thermal safety limit for each predicted future moment point by point, and obtains the thermal safety limit sequence within the preset future time window, providing data basis for subsequent modules.

[0036] Risk Integral Calculation Module 400: Compares the predicted temperature trajectory with the thermal safety upper limit sequence point by point, integrates the portion exceeding the limit over time to obtain the over-temperature risk integral value of each heating device, and then sums the over-temperature risk integral values ​​of each heating device by weight to obtain the total over-temperature risk integral value.

[0037] In the specific implementation process, the predicted temperature trajectory is aligned and compared with the thermal safety upper limit sequence point by point along the time dimension. For each time point, the portion of the predicted temperature exceeding the thermal safety upper limit (the over-temperature amount) is accumulated and integrated to quantify the overall over-temperature risk level within the future time window. The sum of the over-temperature amounts at all time points yields the over-temperature risk integral value of the heating device. The over-temperature risk integral values ​​of multiple heating devices in the system are calculated separately, multiplied by preset device weighting coefficients, and then summed to obtain the total over-temperature risk integral value. The device weighting coefficients are pre-configured based on the criticality of the devices in the system. The total over-temperature risk integral value will serve as the basis for determining the overall over-temperature risk integral value. The data foundation for subsequent modules is based on the classification of equipment weight coefficients according to the severity of equipment failure consequences. For example, there are three levels: Level 1 is critical equipment that causes system downtime or safety accidents, with a weight coefficient of 1.5; Level 2 is important equipment that affects some functions but can be degraded to a lower level, with a weight coefficient of 1.0; and Level 3 is non-critical auxiliary equipment, with a weight coefficient of 0.5. This classification standard is defined by the system safety specification document. For example, in a CNC machine tool system, the spindle servo drive belongs to Level 1 equipment, and the coolant pump motor belongs to Level 3 equipment. Therefore, the former has a weight coefficient of 1.5, and the latter has a weight coefficient of 0.5.

[0038] Energy Efficiency Assessment Module 500: Based on a pre-built first-order thermal response model and candidate fan control actions, it combines the predicted operating load power trajectory, current ambient temperature, and current equipment temperature to deduce the intervention temperature trajectory and calculate the energy efficiency regulation ratio of each candidate fan control action.

[0039] Specifically, in the energy efficiency assessment module, the pre-calibration steps for the pre-calibrated fan and equipment thermal influence weight matrix include: When each heat-generating device is under high load, control all fans to run at the lowest safe speed setting so that all heat-generating devices reach a high temperature steady state. The high temperature steady state is when each heat-generating device is within 80% to 95% of its maximum allowable junction temperature limit, and the temperature change rate does not exceed ±1 degree Celsius per minute for five consecutive minutes. Each fan was individually increased to its preset speed setting, while the remaining fans were kept at their lowest safe speed setting. After the temperature of each heating device has stabilized, record the temperature drop data of each heating device. Based on the maximum temperature drop caused by each fan, the temperature drop data is normalized to obtain the relative cooling weight of each fan for each heat-generating device. Combine all relative cooling weights into a matrix to obtain the fan and equipment thermal influence weight matrix.

[0040] Specifically, in the energy efficiency assessment module, the formula for predicting the intervention temperature trajectory is as follows:

[0041]

[0042] in, For heating equipment Future No. Predicted intervention temperature at each time point For heating equipment thermal time constant, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. The preset predictive control time step, For all fans in the future At what time point did the heating device Overall cooling effect For the pre-calibrated thermal influence weight matrix of fans and equipment, For pre-calibrated fans The cooling efficiency function of rotational speed, The total number of fans, This is due to the delay in cooling response.

[0043] Specifically, in the energy efficiency assessment module, the calculation steps for the energy efficiency regulation ratio of each candidate fan control action include: The predicted intervention temperature trajectory corresponding to the candidate fan control action is compared point by point with the thermal safety upper limit sequence at each moment in the future time window. The part of the temperature exceeding the thermal safety upper limit is accumulated over time to obtain the intervention over-temperature risk integral value of each heat-generating device. The intervention over-temperature risk integral values ​​of all heat-generating devices are weighted and summed to obtain the total intervention over-temperature risk integral value corresponding to the candidate fan control action. Calculate the difference between the over-temperature risk integral value and the total intervention over-temperature risk integral value corresponding to the candidate fan control action to obtain the over-temperature risk integral reduction achieved by the candidate fan control action; Obtain the total fan power consumption increment caused by the candidate fan control action relative to the current fan operating state; Divide the integral reduction in over-temperature risk achieved by the candidate fan control action by the total fan power consumption increment to obtain the energy efficiency regulation ratio of the candidate fan control action.

[0044] Specifically, the candidate fan control action refers to a control combination consisting of one or more fans in their respective selectable gears, and the selectable gear of each fan is not lower than the current operating gear.

[0045] Specifically, in the energy efficiency assessment module, the predicted intervention temperature trajectory is only simulated for equipment with an over-temperature risk integral value greater than zero.

[0046] Specifically, the energy efficiency assessment module is activated only when the total over-temperature risk integral value is greater than zero. If the total over-temperature risk integral value is not greater than zero, the fan control module maintains the current operating speed of each fan unchanged.

[0047] In the specific implementation process, the energy efficiency assessment module 500 first operates each heat-generating device under high-load conditions, which refers to a stable operating state where the device power consumption reaches more than 90% of its rated power. Simultaneously, it controls all fans to operate at the lowest speed setting to ensure that each heat-generating device does not trigger over-temperature protection under this condition. This operation continues until the temperature of each heat-generating device enters the range of 80% to 95% of its maximum allowable junction temperature limit, and the temperature change rate does not exceed ±1 degree Celsius per minute for five consecutive minutes. At this point, it is determined that a high-temperature steady state has been reached. For example, for a servo driver with a maximum allowable junction temperature limit of 125°C, its high-temperature steady-state temperature range is 100°C to 119°C, and the measured stable temperature is 115°C; for a geared motor with a maximum allowable junction temperature limit of 155°C, its high-temperature steady-state temperature range is 124°C to 147°C, and the measured stable temperature is 140°C. Subsequently, the remaining fans are kept in a constant state. The system then sequentially increases each fan's speed to a preset calibration level. Once the system reaches the aforementioned high-temperature steady-state condition again, the temperature drop of each heat-generating device relative to the initial high-temperature steady-state is recorded. For example, when only the first calibration fan is increased to level four, the servo driver temperature drops from 115℃ to 109℃, a temperature drop of 6℃, and the geared motor temperature drops from 140℃ to 138℃, a temperature drop of 2℃. When only the second calibration fan is increased to level four, the servo driver temperature drops by 2.4℃, and the geared motor temperature drops by 5.4℃. Normalization is then performed based on the maximum temperature drop caused by each fan: the weight of the first calibration fan to the servo driver is 6.0 divided by 6.0, which equals 1.0; the weight to the geared motor is 2.0 divided by 6.0, which equals 0.33; the weight of the second calibration fan to the servo driver is 2.4 divided by 5.4, which equals 0.44; and the weight to the geared motor is 5.4 divided by 5.4, which equals 1.0; Finally, the above relative cooling weights are combined into a matrix with fan rows and heat-generating equipment columns to obtain the fan and equipment thermal impact weight matrix; this matrix is ​​used to calculate the total cooling effect of candidate fan control actions on each heat-generating device. At the beginning of each control cycle, it is first determined whether the total over-temperature risk integral value is greater than zero. If it is not greater than zero, subsequent calculations are skipped and the current fan speed is maintained. If it is greater than zero, a candidate fan control action set is generated. The candidate fan control action set includes actions to maintain all current fan speeds and actions for each fan to gradually increase to a higher speed from the current speed. The selectable speed of each fan is not lower than its current operating speed. For each candidate fan control action, using A pre-constructed first-order thermal response model is used to extrapolate the predicted intervention temperature trajectory within a preset time window. This extrapolation requires calculating the sum of the cooling effects generated by each heat-generating device for every candidate fan control action, i.e., the total cooling effect. The calculation of the total cooling effect is based on a piecewise model that considers cooling response delay: in the first few time steps after the start of the control cycle, although the fan has issued an upshift command, the cooling effect has not yet materialized due to motor acceleration, airflow establishment, and thermal conduction inertia; therefore, the total cooling effect is zero. Only after a preset cooling response delay number of steps does the cooling effect reach a stable value. Specifically, the total cooling effect equals the cooling contribution of all fans to the heat-generating device. The contribution of each fan is calculated by multiplying its cooling efficiency value at the target speed by the fan's thermal impact weight on the heat-generating equipment. This weight is obtained through a calibrated fan-equipment thermal impact weight matrix. The cooling efficiency value reflects the maximum temperature drop the fan can provide at a specific speed. This value is obtained through a pre-established lookup table function from speed to cooling temperature drop in a fan speed cooling capacity calibration experiment. Specifically, it is obtained through a pre-calibrated speed cooling efficiency function, which is a multi-dimensional lookup table structure. For example, its inputs include, but are not limited to, the target fan speed, the current ambient temperature range, and the equipment's heat load level. The output is the maximum cooling efficiency the fan can provide under that operating condition. Steady-state cooling temperature drop value; calibration experiments are conducted at three typical ambient temperature points, including but not limited to 20℃, 30℃, and 40℃, and three load levels, including but not limited to low, medium, and high. Each selectable speed of each fan is run individually until thermal steady state, and the equipment temperature drop is recorded to form a three-dimensional data table; during the operation phase, if the current ambient temperature or load is between the calibration points, the cooling efficiency value is calculated using bilinear interpolation; if the requested speed exceeds the calibration range, the value of the closest valid speed is taken and an alarm log is triggered; for example, for any fan at speed three, ambient temperature 26℃, and medium load, the cooling efficiency value is 4.2℃ obtained by looking up the table, and the final value after interpolation is 4.0.05 degrees Celsius. Cooling response delay is an integer parameter representing the number of time steps required from issuing a fan control command to the cooling effect taking full effect. Its value is determined through actual fan start-stop experiments. The number of cooling response delay steps is not a fixed constant, but is dynamically obtained by looking up a table based on the difference between the target fan speed and the current speed. The calibration method is as follows: Under rated load and an ambient temperature of 25°C, control the fan to switch from the current speed to the target speed, and simultaneously record the temperature response curve of the associated heat-generating equipment. Define the moment when the cooling effect takes full effect as the time point when the equipment temperature drops to 95% of the final steady-state temperature drop of this action. Divide this moment by the control cycle and round up to get the number of cooling response delay steps for the corresponding speed switching combination. All possible speed switching combinations are pre-calibrated and stored in the delay lookup table. For example, for any fan switching combination from speed two to speed three, if the time required for the temperature drop to reach 95% of the steady state is measured to be 2.3 seconds, then the number of cooling response delay steps is 3. If switching from speed two to speed four, the delay steps are 4 because the acceleration time is longer. Next, the pre-constructed first-order thermal response model uses a discrete recursive approach for prediction. The recursion starts at the current equipment temperature, and the time step equals the control cycle. After completing the intervention temperature trajectory derivation, this trajectory is compared with the thermal safety upper limit sequence point by point. For the portion exceeding the limit, the over-temperature risk integral value for the heat-generating device is accumulated over time. The weighted sum of the over-temperature risk integral values ​​of all heat-generating devices yields the total over-temperature risk integral value under the candidate fan control action. The difference between the total over-temperature risk integral value and the over-temperature risk integral value is calculated to obtain the over-temperature risk integral reduction. Simultaneously, based on the fan power consumption level table, the total fan power consumption increment caused by the candidate fan control action relative to the current fan state is calculated. The over-temperature risk integral reduction divided by the total fan power consumption increment is the energy efficiency regulation ratio of the candidate fan control action. If the maximum energy efficiency regulation ratio among all candidate fan control actions is lower than the preset energy efficiency threshold, the regulation is abandoned, and the current level is maintained.

[0048] Fan control module 600: Based on the energy efficiency adjustment ratio of each candidate fan control action and the preset energy efficiency threshold, it determines the target fan control action and sends a gear control command to the corresponding target fan.

[0049] Specifically, in the fan control module, after determining the target fan control action, if the energy efficiency regulation ratio corresponding to the target fan control action is lower than the preset energy efficiency threshold, then any fan adjustment will be abandoned, and the current operating speed of each fan will remain unchanged.

[0050] In the specific implementation process, all energy efficiency regulation ratios are compared with the preset energy efficiency threshold. If the energy efficiency regulation ratio of at least one candidate fan control action is greater than or equal to the energy efficiency threshold, the candidate fan control action with the largest energy efficiency regulation ratio is selected as the target fan control action, and the corresponding gear control command is sent to the fan involved in the action. If the energy efficiency regulation ratio of all candidate fan control actions is less than the energy efficiency threshold, no fan adjustment is executed, and the current operating gear of all fans remains unchanged. The energy efficiency threshold is a fixed threshold pre-configured in the system firmware, and its value is determined through energy efficiency calibration experiments. This completes one control cycle of high-temperature prediction-based fan control based on energy efficiency optimization.

[0051] The working principle of the high-temperature predictive start-up fan control system based on energy efficiency optimization provided by this invention is as follows: This invention utilizes a load prediction module 100, based on the predictability of the thermal process, to collect current equipment temperature, ambient temperature, and historical load data within each fixed control cycle, extrapolating the operating load power trajectory for a future period. A temperature prediction module 200, combining a pre-constructed first-order thermal response model with calibrated equipment thermal resistance and thermal time constants, generates a predicted trajectory for future temperature changes. A safety upper limit calculation module 300 calculates the thermal safety upper limit sequence at each moment based on the device's highest junction temperature threshold, safety margin coefficient, and predicted load dynamics. A risk integral calculation module 400 compares and integrates the predicted temperature trajectory with the thermal safety upper limit point by point to quantify the over-temperature risk. If a risk exists, for all candidate fan control actions not lower than the current speed, an energy efficiency assessment module 500 uses the fan and equipment thermal influence weight matrix and cooling benefit model to extrapolate the temperature trajectory after intervention, calculating the ratio of the risk reduction to the increased power consumption for each action, i.e., the energy efficiency regulation ratio. The fan control module 600 ultimately selects the action with the highest energy efficiency regulation ratio that exceeds a preset critical value for execution; otherwise, it maintains the status quo, thereby achieving intelligent closed-loop control that proactively avoids over-temperature risks with minimal heat dissipation energy consumption.

[0052] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0053] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0054] It should also be noted that 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 process, method, article, or apparatus. Unless otherwise specified, 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.

Claims

1. A high-temperature predictive start-up fan control system based on energy efficiency optimization, characterized in that, The fan control system includes: Load prediction module: Based on the operating load power sequence of each heat-generating device within a preset time period in the past, extrapolate to obtain the predicted operating load power trajectory within a preset time window in the future; Temperature prediction module: Based on a pre-built first-order thermal response model, and using the predicted operating load power trajectory, current ambient temperature and current equipment temperature as inputs, it extrapolates the predicted temperature trajectory within a preset time window. Safety limit calculation module: Based on the predicted operating load power trajectory and the maximum allowable junction temperature limit of the device, calculate the sequence of thermal safety limit for each heat-generating device within a preset time window in the future; Risk Integral Calculation Module: The module compares the predicted temperature trajectory with the thermal safety upper limit sequence point by point, integrates the portion exceeding the limit over time, obtains the over-temperature risk integral value of each heating device, and then sums the over-temperature risk integral values ​​of each heating device by weight to obtain the total over-temperature risk integral value. Energy efficiency assessment module: Based on the pre-built first-order thermal response model and candidate fan control actions, it combines the predicted operating load power trajectory, current ambient temperature and current equipment temperature to deduce the intervention temperature trajectory, and calculates the energy efficiency regulation ratio of each candidate fan control action; Fan control module: Based on the energy efficiency adjustment ratio of each candidate fan control action and the preset energy efficiency threshold, determine the target fan control action and send the gear control command to the corresponding target fan.

2. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 1, characterized in that, In the temperature prediction module, the pre-built first-order thermal response model is as follows: in, For heating equipment Future No. The device temperature at each point in time. For heating equipment thermal time constant, The current ambient temperature. For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. This is the preset predictive control time step.

3. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 2, characterized in that, In the safety limit calculation module, the formula for calculating the thermal safety limit sequence of each heat-generating device within a future preset time window is as follows: in, For heating equipment Future No. The thermal safety limit at a given point in time. For heating equipment The device allows for a maximum junction temperature limit. For the temperature rise safety factor, For heating equipment The thermal resistance of the equipment, For heating equipment Future No. The predicted operating load power of the equipment at each point in time.

4. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 3, characterized in that, In the energy efficiency assessment module, the pre-calibration steps for the pre-calibrated fan and equipment thermal impact weight matrix include: When each heat-generating device is under high load, control all fans to run at the lowest safe speed setting so that all heat-generating devices reach a high temperature steady state. The high temperature steady state is when each heat-generating device is within 80% to 95% of its maximum allowable junction temperature limit, and the temperature change rate does not exceed ±1 degree Celsius per minute for five consecutive minutes. Each fan was individually increased to its preset speed setting, while the remaining fans were kept at their lowest safe speed setting. After the temperature of each heating device has stabilized, record the temperature drop data of each heating device. Based on the maximum temperature drop caused by each fan, the temperature drop data is normalized to obtain the relative cooling weight of each fan for each heat-generating device. The fan and equipment thermal influence weight matrix is ​​obtained by combining all the relative cooling weights into a matrix.

5. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 4, characterized in that, In the energy efficiency assessment module, the formula for predicting the intervention temperature trajectory is as follows: in, For heating equipment Future No. Predicted intervention temperature at each time point For heating equipment thermal time constant, For heating equipment Future No. The predicted operating load power of the equipment at each point in time. The preset predictive control time step, For all fans in the future At what time point did the heating device Overall cooling effect For the pre-calibrated thermal influence weight matrix of fans and equipment, For pre-calibrated fans The cooling efficiency function of rotational speed, The total number of fans, This is due to the delay in cooling response.

6. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 5, characterized in that, In the energy efficiency assessment module, the calculation steps for the energy efficiency regulation ratio of each candidate fan control action include: The predicted intervention temperature trajectory corresponding to the candidate fan control action is compared point by point with the thermal safety upper limit sequence at each moment in the future time window. The part of the temperature exceeding the thermal safety upper limit is accumulated over time to obtain the intervention over-temperature risk integral value of each heat-generating device. The intervention over-temperature risk integral values ​​of all heat-generating devices are weighted and summed to obtain the total intervention over-temperature risk integral value corresponding to the candidate fan control action. Calculate the difference between the over-temperature risk integral value and the total intervention over-temperature risk integral value corresponding to the candidate fan control action to obtain the over-temperature risk integral reduction achieved by the candidate fan control action; Obtain the total fan power consumption increment caused by the candidate fan control action relative to the current fan operating state; Divide the integral reduction in over-temperature risk achieved by the candidate fan control action by the total fan power consumption increment to obtain the energy efficiency regulation ratio of the candidate fan control action.

7. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 6, characterized in that, Candidate fan control action refers to a control combination consisting of one or more fans in their respective selectable gears, and the selectable gear of each fan is not lower than the current operating gear.

8. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 7, characterized in that, In the energy efficiency assessment module, only devices with an over-temperature risk integral value greater than zero are used to predict and intervene in temperature trajectories.

9. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 8, characterized in that, The energy efficiency assessment module is activated only when the total over-temperature risk integral value is greater than zero. If the total over-temperature risk integral value is not greater than zero, the fan control module will maintain the current operating speed of each fan.

10. The high-temperature predictive start-up fan control system based on energy efficiency optimization according to claim 9, characterized in that, In the fan control module, after determining the target fan control action, if the energy efficiency regulation ratio corresponding to the target fan control action is lower than the preset energy efficiency threshold, then any fan adjustment will be abandoned, and the current operating speed of each fan will remain unchanged.

Images