Method and system for adjusting power of energy consumption equipment

Abstract

The invention provides an energy consumption equipment power regulation method and system, and the method comprises the steps: obtaining the power time sequence data of pre-regulation energy consumption equipment in a target building area, carrying out the cost function construction step of each piece of pre-regulation energy consumption equipment power time sequence data, and obtaining a composite cost function of each piece of pre-regulation energy consumption equipment; based on the composite cost function, optimization is carried out under the target total power regulation quantity and the rated working power to obtain the optimal power down-regulation quantity; generating a regulation instruction according to the optimal power down-regulation amount; the cost function construction step comprises the following steps: extracting average power, power variance and power change rate based on the power time sequence data of the pre-regulation energy consumption equipment; obtaining a device adjustment weight corresponding to the pre-adjustment energy consumption device; generating an operation penalty term based on the average power, the power variance and the power change rate; and generating a power adjustment cost item based on the equipment adjustment weight, and integrating the operation penalty item to construct a pre-adjustment energy consumption equipment composite cost function. According to the invention, the stability of cooperative power regulation of the multi-energy-consumption equipment can be improved.

Description

Technical Field

[0001] This invention belongs to the field of building energy management technology, and in particular relates to a method and system for regulating the power of energy-consuming equipment. Background Technology

[0002] In the field of building energy management technology, to reduce emissions and save energy, it is often necessary to coordinate the power regulation of multiple energy-consuming devices in a building, such as air conditioners, fans, and lighting. Traditional solutions typically allocate the target total power regulation amount based on the rated power of each energy-consuming device or a preset priority. However, in actual operation, the real-time operating status of similar energy-consuming devices may differ, and some energy-consuming devices may be operating under high fluctuations or high loads. If they are allocated the same power regulation amount as other similar energy-consuming devices, the actual power regulation effect will fluctuate significantly and the reliability will be low.

[0003] To address the aforementioned issues, existing technologies collect multi-source data, including equipment operating status, utilize machine learning models for energy consumption prediction, and employ reinforcement learning to generate power regulation control strategies. This approach adjusts the power regulation control strategy based on real-time equipment operating data, improving the stability of actual power regulation to some extent. However, the reinforcement learning reward mechanism of this existing technology is designed for continuous improvement based on whether actual energy consumption is lower than or close to the predicted value. Therefore, when applied to scenarios requiring rapid achievement of a given target total power regulation, it may repeatedly apply actions to a few easily adjustable and fast-responding devices, overburdening some devices and easily leading to inaccurate power regulation allocation and significant power fluctuations. Furthermore, due to the inherent characteristics of reinforcement learning, the aforementioned power regulation actions do not guarantee achieving the total power regulation target. Therefore, existing technologies cannot guarantee achieving the target total power regulation and are prone to inaccurate power regulation allocation and large power fluctuations, significantly reducing the stability of coordinated power regulation across multiple energy-consuming devices. Summary of the Invention

[0004] The present invention aims to provide a method and system for regulating the power of energy-consuming equipment to solve the above-mentioned technical problems. Improve the stability of coordinated power regulation for multi-energy-consuming devices.

[0005] To address the aforementioned technical problems, this invention provides a method for regulating the power of energy-consuming equipment, comprising the following steps: Obtain the power time series data of several pre-adjustable energy consumption devices within the target building area, and perform the cost function construction step on the power time series data of each pre-adjustable energy consumption device to obtain the composite cost function corresponding to each pre-adjustable energy consumption device; Based on the composite cost function, the optimal power reduction amount corresponding to each pre-adjusted energy consumption device is obtained by optimizing the solution under the preset target total power adjustment amount and the preset rated working power. An adjustment command is generated based on the optimal power reduction amount; the adjustment command is used to control each pre-adjusted energy consumption device to adjust its power according to the optimal power reduction amount. The cost function construction steps include: Based on the power time-series data of pre-regulated energy consumption equipment, the average power, power variance, and power change rate are extracted. Obtain the equipment adjustment weight corresponding to the pre-adjusted energy consumption equipment under the preset equipment adjustment weight sequence; Based on the average power, the power variance, and the power change rate, an operational penalty term is generated under a preset penalty weight. Based on the device adjustment weights, the power reduction amount to be obtained is weighted to generate a power adjustment cost item; By integrating the power regulation cost term and the operation penalty term, a composite cost function corresponding to the pre-regulated energy consumption equipment is constructed.

[0006] The above scheme quantifies the adjustment cost of an energy-consuming device under a power reduction by constructing a power adjustment cost term through weighted calculation of the power reduction amount to be obtained. Furthermore, the operating penalty term, constructed based on the average power, power variance, and power change rate reflecting the real-time operating status of the device, reflects the sensitivity of the current energy-consuming device to power adjustment actions. A larger penalty value indicates greater power fluctuations caused by power adjustment. The composite cost function constructed based on the operating penalty term and the power adjustment cost term is optimized under a preset target total power adjustment amount and a preset rated operating power. This ensures that the target total power adjustment amount is achieved while allocating optimal power reduction amounts to each energy-consuming device that do not cause significant power fluctuations. This avoids large-scale power adjustments to energy-consuming devices with high sensitivity to power adjustment actions, thus suppressing power fluctuations caused by coordinated power adjustment of multiple energy-consuming devices and improving the stability of coordinated power adjustment of multiple devices.

[0007] Furthermore, the optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device includes: integrating the composite cost functions corresponding to several pre-adjusted energy consumption devices to obtain a total adjustment cost function; constructing a total adjustment amount constraint based on the preset target total power adjustment amount; constructing a minimum adjustment constraint for a single device based on the preset rated operating power; and optimizing the solution under the total adjustment amount constraint and the minimum adjustment constraint for a single device with minimizing the total adjustment cost function as the optimization objective to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

[0008] Furthermore, the step of minimizing the total adjustment cost function as the optimization objective, and performing optimization under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, includes: constructing a constrained quadratic optimization problem model based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment per device constraint; and performing SLSQP solution based on the constrained quadratic optimization problem model to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption device.

[0009] In the above scheme, the total adjustment constraint constructed based on the target total power adjustment amount can restrict the sum of the power reduction amounts allocated to all energy-consuming devices to reach the target total power adjustment amount. The minimum adjustment constraint for a single device constructed based on the preset rated operating power can limit the power reduction amount allocated to a single energy-consuming device to prevent it from exceeding the power adjustment range of that energy-consuming device, ensuring the feasibility of the subsequently obtained optimal power reduction amount. Moreover, under the premise of satisfying the total adjustment constraint and the minimum adjustment constraint for a single device, this scheme performs optimization under the total adjustment cost function to find a set of optimal power reduction amounts with the minimum total adjustment cost function. Therefore, it can ensure that the cost required for each energy-consuming device to adjust its power according to the optimal power reduction amount is minimized, reducing the energy consumption of the overall power adjustment. Compared with the existing technology that generates power adjustment strategies based on reinforcement learning, the multi-energy-consuming device collaborative power adjustment of this scheme has higher stability.

[0010] Furthermore, after generating the adjustment instruction based on the optimal power reduction amount, the method further includes: obtaining the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power based on the optimal power reduction amount; obtaining the power adjustment deviation based on the actual power reduction amount and the optimal power reduction amount, and generating structured prompt words based on the power adjustment deviation; and performing retrieval enhancement generation capability analysis under a preset large language model based on the structured prompt words, and updating the device adjustment weight sequence.

[0011] In the above scheme, after the pre-adjusted energy-consuming devices perform power adjustment based on the optimal power reduction amount, the actual power reduction amount of each pre-adjusted energy-consuming device is obtained. The power adjustment deviation obtained by comparing the actual power reduction amount with the optimal power reduction amount is compared, and the device adjustment weight sequence is updated based on the power adjustment deviation. Thus, the updated device adjustment weight sequence can more accurately reflect the controllability of each energy-consuming device under the current operating state. When performing power adjustment based on the updated device adjustment weight sequence, the problem of deviation between the actual power reduction amount and the optimal power reduction amount caused by changes in the operating state of each energy-consuming device can be mitigated, thereby improving the adjustment efficiency of subsequent coordinated power adjustment of multiple energy-consuming devices.

[0012] Further, the step of extracting average power, power variance, and power change rate based on the power time-series data of the pre-adjusted energy consumption device includes: performing mean processing on the power time-series data to obtain average power; performing difference processing on the power time-series data and the average power to obtain power difference time-series data, and performing mean processing on the power difference time-series data to obtain power variance; and obtaining the power change rate based on the power time-series data at a preset interval.

[0013] Further, the step of generating an operational penalty item based on the average power, the power variance, and the power change rate under a preset penalty weight includes: generating an energy efficiency penalty item based on the average power under a high power consumption weight with a preset penalty weight; generating a fluctuation penalty item based on the power variance under a power consumption fluctuation weight with a preset penalty weight; generating a sudden change penalty item based on the power change rate under a power speed change weight with a preset penalty weight; and integrating the energy efficiency penalty item, the fluctuation penalty item, and the sudden change penalty item to generate an operational penalty item.

[0014] In the above scheme, the energy efficiency penalty term generated based on average power characterizes the baseline energy consumption level of the energy-consuming equipment. A larger energy efficiency penalty term results in greater energy consumption and a higher risk of power fluctuations when the power of the equipment is reduced further. The fluctuation penalty term generated based on power variance characterizes the degree of power fluctuation during operation. A larger fluctuation penalty term results in greater power fluctuations when the power of the equipment is reduced further. The abrupt change penalty term generated based on the power change rate penalizes rapid power changes. A larger abrupt change penalty term results in greater power fluctuations when the power is reduced. The operating penalty term generated by this scheme based on the above energy efficiency penalty term, fluctuation penalty term, and abrupt change penalty term can better guide subsequent optimization solutions to avoid excessive power reduction of high-energy-consuming or highly volatile equipment, thereby improving the stability of coordinated power regulation of multiple energy-consuming devices.

[0015] This invention also provides an energy-consuming equipment power regulation system for implementing any of the above-described energy-consuming equipment power regulation methods, comprising: a composite cost function module for acquiring power time-series data of several pre-regulated energy-consuming equipment within a target building area, and performing a cost function construction step on the power time-series data of each pre-regulated energy-consuming equipment to obtain a composite cost function corresponding to each pre-regulated energy-consuming equipment; a power reduction amount acquisition module for optimizing the solution based on the composite cost function under a preset target total power regulation amount and a preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-regulated energy-consuming equipment; and a power regulation module for generating regulation commands based on the optimal power reduction amount. The adjustment command is used to control each pre-adjusted energy consumption device to adjust its power according to the optimal power reduction amount; the cost function construction step includes: extracting average power, power variance, and power change rate based on the power time series data of the pre-adjusted energy consumption device; obtaining the device adjustment weight corresponding to the pre-adjusted energy consumption device under a preset device adjustment weight sequence; generating an operating penalty term based on the average power, the power variance, and the power change rate under a preset penalty weight; weighting the power reduction amount to be obtained based on the device adjustment weight to generate a power adjustment cost term; and integrating the power adjustment cost term and the operating penalty term to construct a composite cost function corresponding to the pre-adjusted energy consumption device.

[0016] Furthermore, the optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device includes: integrating the composite cost functions corresponding to several pre-adjusted energy consumption devices to obtain a total adjustment cost function; constructing a total adjustment amount constraint based on the preset target total power adjustment amount; constructing a minimum adjustment constraint for a single device based on the preset rated operating power; and optimizing the solution under the total adjustment amount constraint and the minimum adjustment constraint for a single device with minimizing the total adjustment cost function as the optimization objective to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

[0017] Furthermore, the step of minimizing the total adjustment cost function as the optimization objective, and performing optimization under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, includes: constructing a constrained quadratic optimization problem model based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment per device constraint; and performing SLSQP solution based on the constrained quadratic optimization problem model to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption device.

[0018] Furthermore, after generating the adjustment instruction based on the optimal power reduction amount, the method further includes: obtaining the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power based on the optimal power reduction amount; obtaining the power adjustment deviation based on the actual power reduction amount and the optimal power reduction amount, and generating structured prompt words based on the power adjustment deviation; and performing retrieval enhancement generation capability analysis under a preset large language model based on the structured prompt words, and updating the device adjustment weight sequence.

[0019] The power regulation cost term constructed by the above scheme can quantify the regulation cost of the energy-consuming equipment under a power reduction, and the constructed operating penalty term can reflect the sensitivity of the current energy-consuming equipment to power regulation actions. Based on the above operating penalty term and power regulation cost term, the composite cost function is optimized under the preset target total power regulation amount and preset rated operating power. It can ensure that the target total power regulation amount is achieved while avoiding large power regulation of energy-consuming equipment with high sensitivity to power regulation actions. Overall, it suppresses the power fluctuations generated when multiple energy-consuming equipment coordinate power regulation and improves the stability of multi-equipment coordinated power regulation. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the steps of a power regulation method for energy-consuming equipment provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an energy-consuming equipment power regulation system provided in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 This embodiment provides a method for regulating the power of energy-consuming equipment, including the following steps: Step S1: Obtain the power time series data of several pre-adjustable energy consumption devices within the target building area, and perform the cost function construction step on the power time series data of each pre-adjustable energy consumption device to obtain the composite cost function corresponding to each pre-adjustable energy consumption device; Step S2: Based on the composite cost function, perform optimization solution under the preset target total power adjustment amount and preset rated working power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device; Step S3: Generate an adjustment command based on the optimal power reduction amount; the adjustment command is used to control each pre-adjusted energy consumption device to adjust its power according to the optimal power reduction amount; The cost function construction steps include: Based on the power time-series data of pre-regulated energy consumption equipment, the average power, power variance, and power change rate are extracted. Obtain the equipment adjustment weight corresponding to the pre-adjusted energy consumption equipment under the preset equipment adjustment weight sequence; Based on the average power, the power variance, and the power change rate, an operational penalty term is generated under a preset penalty weight. Based on the device adjustment weights, the power reduction amount to be obtained is weighted to generate a power adjustment cost item; By integrating the power regulation cost term and the operation penalty term, a composite cost function corresponding to the pre-regulated energy consumption equipment is constructed.

[0023] The above embodiments construct a power adjustment cost term by weighting the power reduction amount to be obtained through device adjustment weights. This can quantify the adjustment cost of the energy-consuming device under a power reduction amount. Furthermore, the operation penalty term, constructed based on the average power, power variance, and power change rate reflecting the real-time operating status of the device, can reflect the sensitivity of the current energy-consuming device to power adjustment actions. If the penalty value of the operation penalty term is larger, the power fluctuation caused by the current energy-consuming device's power adjustment will be greater. The composite cost function constructed based on the above operation penalty term and power adjustment cost term is optimized under the preset target total power adjustment amount and preset rated operating power. This can ensure that the target total power adjustment amount is achieved while allocating the optimal power reduction amount to each energy-consuming device without causing large power fluctuations. This avoids the situation where energy-consuming devices with high sensitivity to power adjustment actions are subject to large power adjustments, thereby suppressing the power fluctuations generated when multiple energy-consuming devices coordinate power adjustment and improving the stability of multi-device coordinated power adjustment.

[0024] Furthermore, the optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device includes: integrating the composite cost functions corresponding to several pre-adjusted energy consumption devices to obtain a total adjustment cost function; constructing a total adjustment amount constraint based on the preset target total power adjustment amount; constructing a minimum adjustment constraint for a single device based on the preset rated operating power; and optimizing the solution under the total adjustment amount constraint and the minimum adjustment constraint for a single device with minimizing the total adjustment cost function as the optimization objective to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

[0025] Furthermore, the step of minimizing the total adjustment cost function as the optimization objective, and performing optimization under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, includes: constructing a constrained quadratic optimization problem model based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment per device constraint; and performing SLSQP solution based on the constrained quadratic optimization problem model to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption device.

[0026] In the above embodiments, the total adjustment constraint constructed based on the target total power adjustment amount can restrict the sum of the power reduction amounts allocated to all energy-consuming devices to reach the target total power adjustment amount; the minimum adjustment constraint for a single device constructed based on the preset rated operating power can limit the power reduction amount allocated to a single energy-consuming device to prevent it from exceeding the power adjustment range of that energy-consuming device, ensuring the feasibility of the subsequently obtained optimal power reduction amount. Moreover, this embodiment optimizes the solution based on the total adjustment cost function under the premise of satisfying the total adjustment constraint and the minimum adjustment constraint for a single device, in order to find a set of optimal power reduction amounts with the minimum total adjustment cost function. Therefore, it can ensure that the cost required for each energy-consuming device to adjust its power according to the optimal power reduction amount is minimized, reducing the energy consumption of the overall power adjustment. Compared with the existing technology that generates power adjustment strategies based on reinforcement learning, the multi-energy-consuming device collaborative power adjustment in this embodiment has higher stability.

[0027] Furthermore, after generating the adjustment instruction based on the optimal power reduction amount, the method further includes: obtaining the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power based on the optimal power reduction amount; obtaining the power adjustment deviation based on the actual power reduction amount and the optimal power reduction amount, and generating structured prompt words based on the power adjustment deviation; and performing retrieval enhancement generation capability analysis under a preset large language model based on the structured prompt words, and updating the device adjustment weight sequence.

[0028] In the above embodiments, after the pre-adjusted energy-consuming devices perform power adjustment based on the optimal power reduction amount, the actual power reduction amount of each pre-adjusted energy-consuming device is obtained. The power adjustment deviation obtained by comparing the actual power reduction amount with the optimal power reduction amount is compared, and the device adjustment weight sequence is updated in the subsequent process based on the power adjustment deviation. Thus, the updated device adjustment weight sequence can more accurately reflect the controllability of each energy-consuming device in the current operating state. When performing power adjustment in the subsequent process based on the updated device adjustment weight sequence, the problem of deviation between the actual power reduction amount and the optimal power reduction amount caused by changes in the operating state of each energy-consuming device can be mitigated, thereby improving the adjustment efficiency of subsequent coordinated power adjustment of multiple energy-consuming devices.

[0029] For example, and not as a limitation, consider the following energy-consuming devices within a target building area: Air Conditioner 1, Air Conditioner 2, and Air Conditioner 3. Air Conditioner 1 and Air Conditioner 3 are pre-adjusted energy-consuming devices. At the current time of 10:00, the set target total power adjustment is 1000W. At this time, the current power of Air Conditioner 1 is 1475W, the current power of Air Conditioner 2 is 1719W, and the current power of Air Conditioner 3 is 598W. After executing the energy-consuming device power adjustment method provided in the above embodiment, the optimal power reduction for Air Conditioner 1 is 646W, and the optimal power reduction for Air Conditioner 3 is 354W. After Air Conditioner 1 and Air Conditioner 3 adjust their power according to their corresponding optimal power reduction amounts, the actual power reduction for Air Conditioner 1 is 1210W, and the actual power reduction for Air Conditioner 3 is 0W. Since Air Conditioner 2 is not a pre-adjusted energy-consuming device at this time, both the optimal power reduction and the actual power reduction for Air Conditioner 2 are 0W. In the original equipment adjustment weight sequence, the adjustment weight of air conditioner 1 is 1.0, the adjustment weight of air conditioner 2 is 2.0, and the adjustment weight of air conditioner 3 is 1.0. After executing the energy consumption equipment power adjustment method provided in the above embodiment, the adjustment weight of air conditioner 1 in the updated equipment adjustment weight sequence is 0.4, the adjustment weight of air conditioner 2 is 1.0, and the adjustment weight of air conditioner 3 is 2.5.

[0030] It is evident that for air conditioner 1, whose actual power reduction far exceeds the optimal power reduction, it responds positively to the optimal power reduction, achieving effective power regulation. Therefore, the regulation weight of air conditioner 1 in the equipment regulation weight sequence is reduced, so that more power reduction will be allocated to air conditioner 1 in subsequent power regulation. However, due to the penalty limit of the operating penalty term, the power reduction allocated to air conditioner 1 will also be constrained to a certain extent to prevent it from exceeding the appropriate range. Furthermore, for air conditioner 3, whose actual power reduction is far lower than the optimal power reduction, it responds negatively to the optimal power reduction. Therefore, it is considered that regulating the power of air conditioner 3 requires a high cost. Thus, the regulation weight of air conditioner 3 in the equipment regulation weight sequence is increased to reduce or even avoid allocating a large amount of power reduction from air conditioner 3; and the regulation weight of air conditioner 2 is reduced to offset the sharp increase in the regulation weight of air conditioner 3.

[0031] For example, and not as a limitation, consider the following energy-consuming devices within a target building area: Air Conditioner 1, Air Conditioner 2, and Air Conditioner 3. Air Conditioner 1 and Air Conditioner 3 are pre-adjusted energy-consuming devices. At the current time of 15:00, the set target total power adjustment is 1500W. At this time, the current power of Air Conditioner 1 is 1759W, the current power of Air Conditioner 2 is 1879W, and the current power of Air Conditioner 3 is 570W. After executing the energy-consuming device power adjustment method provided in the above embodiment, the optimal power reduction for Air Conditioner 1 is 948W, and the optimal power reduction for Air Conditioner 3 is 552W. After Air Conditioner 1 and Air Conditioner 3 adjust their power according to their corresponding optimal power reduction amounts, the actual power reduction for Air Conditioner 1 is 1686W, and the actual power reduction for Air Conditioner 3 is 142W. Since Air Conditioner 2 is not a pre-adjusted energy-consuming device at this time, both the optimal power reduction and the actual power reduction for Air Conditioner 2 are 0W. In the last updated equipment adjustment weight sequence, the corresponding adjustment weights of air conditioner 1, air conditioner 2, and air conditioner 3 were 0.4, 1.0, and 2.5, respectively. After executing the energy consumption equipment power adjustment method provided in the above embodiment, the updated equipment adjustment weight sequence has an adjustment weight of 0.3 for air conditioner 1, 1.0 for air conditioner 2, and 2.0 for air conditioner 3.

[0032] It is evident that Air Conditioner 1 responded significantly more actively than the optimal power reduction in both the first power adjustment at 10:00 and the second power adjustment at 15:00. Therefore, the adjustment weight of Air Conditioner 1 in the equipment adjustment weight sequence is further reduced, so that more power reduction will be allocated to Air Conditioner 1 in subsequent power adjustments. Air Conditioner 3 did not respond in the first power adjustment at 10:00, resulting in an actual power reduction far lower than the optimal power reduction. Therefore, the adjustment weight of Air Conditioner 3 in the first power adjustment increased from 1.0 to 2.5. In the second power adjustment at 15:00, although the actual power reduction of Air Conditioner 3 was still lower than the optimal power reduction, it had already generated some response, and the deviation between its actual power reduction and the optimal power reduction decreased. This indicates that the cost of adjusting the power of Air Conditioner 3 is reduced, and more power reduction can be allocated to it. Therefore, the adjustment weight of Air Conditioner 3 in the equipment adjustment weight sequence is updated from 2.5 to 2.0.

[0033] As can be seen from the above examples, the above embodiments can adapt to situations where there is a dynamic deviation between the actual power reduction amount and the optimal power reduction amount due to changes in the operating status of each energy-consuming device. This improves the matching device adjustment weight for each energy-consuming device and enhances the adjustment efficiency of subsequent coordinated power adjustment of multiple energy-consuming devices.

[0034] Further, the step of extracting average power, power variance, and power change rate based on the power time-series data of the pre-adjusted energy consumption device includes: performing mean processing on the power time-series data to obtain average power; performing difference processing on the power time-series data and the average power to obtain power difference time-series data, and performing mean processing on the power difference time-series data to obtain power variance; and obtaining the power change rate based on the power time-series data at a preset interval.

[0035] Further, the step of generating an operational penalty item based on the average power, the power variance, and the power change rate under a preset penalty weight includes: generating an energy efficiency penalty item based on the average power under a high power consumption weight with a preset penalty weight; generating a fluctuation penalty item based on the power variance under a power consumption fluctuation weight with a preset penalty weight; generating a sudden change penalty item based on the power change rate under a power speed change weight with a preset penalty weight; and integrating the energy efficiency penalty item, the fluctuation penalty item, and the sudden change penalty item to generate an operational penalty item.

[0036] In the above embodiments, the energy efficiency penalty term generated based on the average power can characterize the baseline energy consumption level of the energy-consuming device. A larger energy efficiency penalty term results in more energy consumption and a greater risk of power fluctuations when the power of the device is reduced further. The fluctuation penalty term generated based on the power variance can characterize the degree of power fluctuation during operation. A larger fluctuation penalty term results in greater power fluctuations when the power of the device is reduced further. The abrupt change penalty term generated based on the power change rate can penalize rapid power changes. A larger abrupt change penalty term results in greater power fluctuations when the power is reduced. The operating penalty term generated in this embodiment based on the above energy efficiency penalty term, fluctuation penalty term, and abrupt change penalty term can better guide subsequent optimization solutions to avoid excessive power reduction of high-energy-consuming or highly volatile energy-consuming devices, thereby improving the stability of coordinated power regulation of multiple energy-consuming devices.

[0037] In one embodiment, the HLW8032 energy monitoring module, which integrates an ESP32 microcontroller, collects power time-series data in real time over a preset time period, wherein the current time is calculated based on the power time-series data. The The formula for the average power of an energy-consuming device is as follows: ; in, The first in the power timing data The energy-consuming device in the preset time period Operating power corresponding to each time period Current time The The average power of each energy-consuming device The preset time period is the length of the time.

[0038] Calculate the current time based on power timing data The The formula for the power variance of an energy-consuming device is as follows: ; in, Current time The The power variance of each energy-consuming device.

[0039] Calculate the current time based on power timing data The The formula for the power change rate of an energy-consuming device is as follows: ; in, Current time The Power change rate of each energy-consuming device; Current time The The power corresponding to each energy-consuming device; The time interval for calculating the rate of change of power; The previous time interval Corresponding time The The power corresponding to each energy-consuming device.

[0040] In the above embodiments, based on the current time The The average power, power variance, and power change rate of each energy-consuming device are used to generate an operating penalty term under a preset penalty weight, and the current time is constructed based on the device adjustment weight. The The composite cost function for each energy-consuming device is as follows: ; in, Current time The The composite cost function of an energy-consuming device For the first The equipment adjustment weight of each energy-consuming device For high power consumption weighting, As a weight for power consumption fluctuation, For power-speed conversion weights, To obtain the first Power reduction of individual energy-consuming devices For power regulation cost, , As an energy efficiency penalty item, For fluctuation penalty items, This is a mutation penalty term.

[0041] In one embodiment, the composite cost function corresponding to a plurality of pre-adjusted energy consumption devices is used to obtain the total adjustment cost function as follows: ; in, This represents the total number of pre-regulated energy-consuming devices. Current time The total adjustment cost function.

[0042] The above embodiments take minimizing the total adjustment cost function as the optimization objective, that is, the optimization objective is... And the total adjustment constraint is: ; in, The target total power adjustment amount. The minimum adjustment constraint for a single device constructed in this embodiment is: ; in, This is the rated operating power.

[0043] In the above embodiments, based on the total adjustment cost function The total adjustment constraint and the minimum adjustment constraint for a single device are used to construct a constrained quadratic optimization problem model. Based on the constrained quadratic optimization problem model, SLSQP is used to solve the problem and obtain the optimal power reduction amount corresponding to the pre-regulated energy consumption device.

[0044] Please see Figure 2The present invention also provides an energy-consuming equipment power adjustment system for implementing any of the above-described energy-consuming equipment power adjustment methods, comprising: a composite cost function module for acquiring power time-series data of several pre-adjustable energy-consuming equipment within a target building area, and performing a cost function construction step on the power time-series data of each pre-adjustable energy-consuming equipment to obtain a composite cost function corresponding to each pre-adjustable energy-consuming equipment; a power reduction amount acquisition module for optimizing the solution based on the composite cost function under a preset target total power adjustment amount and a preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-adjustable energy-consuming equipment; and a power adjustment module for generating an adjustment command based on the optimal power reduction amount. The adjustment command is used to control each pre-adjusted energy consumption device to adjust its power according to the optimal power reduction amount; the cost function construction step includes: extracting average power, power variance, and power change rate based on the power time series data of the pre-adjusted energy consumption device; obtaining the device adjustment weight corresponding to the pre-adjusted energy consumption device under a preset device adjustment weight sequence; generating an operating penalty term based on the average power, the power variance, and the power change rate under a preset penalty weight; weighting the power reduction amount to be obtained based on the device adjustment weight to generate a power adjustment cost term; and integrating the power adjustment cost term and the operating penalty term to construct a composite cost function corresponding to the pre-adjusted energy consumption device.

[0045] Furthermore, the optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device includes: integrating the composite cost functions corresponding to several pre-adjusted energy consumption devices to obtain a total adjustment cost function; constructing a total adjustment amount constraint based on the preset target total power adjustment amount; constructing a minimum adjustment constraint for a single device based on the preset rated operating power; and optimizing the solution under the total adjustment amount constraint and the minimum adjustment constraint for a single device with minimizing the total adjustment cost function as the optimization objective to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

[0046] Furthermore, the step of minimizing the total adjustment cost function as the optimization objective, and performing optimization under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, includes: constructing a constrained quadratic optimization problem model based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment per device constraint; and performing SLSQP solution based on the constrained quadratic optimization problem model to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption device.

[0047] Furthermore, after generating the adjustment instruction based on the optimal power reduction amount, the method further includes: obtaining the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power based on the optimal power reduction amount; obtaining the power adjustment deviation based on the actual power reduction amount and the optimal power reduction amount, and generating structured prompt words based on the power adjustment deviation; and performing retrieval enhancement generation capability analysis under a preset large language model based on the structured prompt words, and updating the device adjustment weight sequence.

[0048] The power regulation cost term constructed in the above embodiments can quantify the regulation cost of the energy-consuming device under a power reduction amount, and the constructed operating penalty term can reflect the sensitivity of the current energy-consuming device to power regulation actions. Based on the composite cost function constructed by the above operating penalty term and power regulation cost term, optimization is performed under the preset target total power regulation amount and preset rated operating power. This can ensure that the target total power regulation amount is achieved while avoiding large-scale power regulation of energy-consuming devices with high sensitivity to power regulation actions. Overall, it suppresses the power fluctuations generated when multiple energy-consuming devices coordinate power regulation and improves the stability of multi-device coordinated power regulation.

[0049] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for regulating the power of energy-consuming equipment, characterized in that, Includes the following steps: Obtain the power time series data of several pre-adjustable energy consumption devices within the target building area, and perform the cost function construction step on the power time series data of each pre-adjustable energy consumption device to obtain the composite cost function corresponding to each pre-adjustable energy consumption device; Based on the composite cost function, the optimal power reduction amount corresponding to each pre-adjusted energy consumption device is obtained by optimizing the solution under the preset target total power adjustment amount and the preset rated working power. An adjustment command is generated based on the optimal power reduction amount; The adjustment command is used to control each pre-adjusted energy consumption device to adjust its power according to the optimal power reduction amount. The cost function construction steps include: Based on the power time-series data of pre-regulated energy consumption equipment, the average power, power variance, and power change rate are extracted. Obtain the equipment adjustment weight corresponding to the pre-adjusted energy consumption equipment under the preset equipment adjustment weight sequence; Based on the average power, the power variance, and the power change rate, an operational penalty term is generated under a preset penalty weight. Based on the device adjustment weights, the power reduction amount to be obtained is weighted to generate a power adjustment cost item; By integrating the power regulation cost term and the operation penalty term, a composite cost function corresponding to the pre-regulated energy consumption equipment is constructed.

2. The power regulation method for energy-consuming equipment as described in claim 1, characterized in that, The optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power is used to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, including: The composite cost functions corresponding to several pre-regulated energy consumption devices are integrated to obtain the total regulation cost function; Based on the preset target total power adjustment amount, construct the total adjustment amount constraint; Based on the preset rated operating power, construct the minimum adjustment constraint for a single device; With minimizing the total adjustment cost function as the optimization objective, optimization is performed under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

3. The power regulation method for energy-consuming equipment as described in claim 2, characterized in that, The optimization objective is to minimize the total adjustment cost function. The optimization is performed under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount for each pre-adjusted energy consumption device, including: Based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment constraint for a single device, a constrained quadratic optimization problem model is constructed. Based on the constrained quadratic optimization problem model, the SLSQP solution is performed to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption equipment.

4. The power regulation method for energy-consuming equipment as described in claim 1, characterized in that, After generating the adjustment command based on the optimal power reduction amount, the method further includes: Obtain the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power according to the optimal power reduction amount; Based on the actual power reduction amount and the optimal power reduction amount, the power adjustment deviation is obtained, and structured prompt words are generated according to the power adjustment deviation; Based on the structured prompts, the retrieval enhancement generation capability is analyzed under a preset large language model, and the device adjustment weight sequence is updated.

5. The power regulation method for energy-consuming equipment as described in claim 1, characterized in that, The power time-series data based on pre-regulated energy consumption equipment is used to extract average power, power variance, and power change rate, including: The average power is obtained by averaging the power time-series data. The power time series data and the average power are processed by difference to obtain power difference time series data, and the power variance is obtained by mean processing based on the power difference time series data. Based on the power time-series data, the power change rate is obtained at a preset interval.

6. The power regulation method for energy-consuming equipment as described in claim 5, characterized in that, The step of generating a running penalty term based on the average power, the power variance, and the power change rate under a preset penalty weight includes: Based on the average power, an energy efficiency penalty term is generated under a high power consumption weight with a preset penalty weight; Based on the power variance, a fluctuation penalty term is generated under a preset penalty weight for power fluctuation. Based on the power change rate, a mutation penalty term is generated under the power speed change weight with a preset penalty weight; The energy efficiency penalty, fluctuation penalty, and mutation penalty are integrated to generate an operational penalty.

7. A power regulation system for energy-consuming equipment, characterized in that, A method for regulating the power of an energy-consuming device as described in any one of claims 1 to 6, comprising: The composite cost function module is used to obtain the power time series data of several pre-adjustable energy consumption devices within the target building area, and to perform the cost function construction step on the power time series data of each pre-adjustable energy consumption device to obtain the composite cost function corresponding to each pre-adjustable energy consumption device. The power reduction acquisition module is used to perform optimization solutions based on the composite cost function under the preset target total power adjustment amount and the preset rated working power to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device. A power regulation module is used to generate a regulation command based on the optimal power reduction amount; the regulation command is used to control each pre-regulated energy consumption device to perform power regulation based on the optimal power reduction amount. The cost function construction steps include: Based on the power time-series data of pre-regulated energy consumption equipment, the average power, power variance, and power change rate are extracted. Obtain the equipment adjustment weight corresponding to the pre-adjusted energy consumption equipment under the preset equipment adjustment weight sequence; Based on the average power, the power variance, and the power change rate, an operational penalty term is generated under a preset penalty weight. Based on the device adjustment weights, the power reduction amount to be obtained is weighted to generate a power adjustment cost item; By integrating the power regulation cost term and the operation penalty term, a composite cost function corresponding to the pre-regulated energy consumption equipment is constructed.

8. The power regulation system for energy-consuming equipment as described in claim 7, characterized in that, The optimization solution based on the composite cost function under the preset target total power adjustment amount and the preset rated operating power is used to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device, including: The composite cost functions corresponding to several pre-regulated energy consumption devices are integrated to obtain the total regulation cost function; Based on the preset target total power adjustment amount, construct the total adjustment amount constraint; Based on the preset rated operating power, construct the minimum adjustment constraint for a single device; With minimizing the total adjustment cost function as the optimization objective, optimization is performed under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount corresponding to each pre-adjusted energy consumption device.

9. The power regulation system for energy-consuming equipment as described in claim 8, characterized in that, The optimization objective is to minimize the total adjustment cost function. The optimization is performed under the constraints of total adjustment amount and minimum adjustment per device to obtain the optimal power reduction amount for each pre-adjusted energy consumption device, including: Based on the total adjustment cost function, the total adjustment amount constraint, and the minimum adjustment constraint for a single device, a constrained quadratic optimization problem model is constructed. Based on the constrained quadratic optimization problem model, the SLSQP solution is performed to obtain the optimal power reduction amount corresponding to the pre-adjusted energy consumption equipment.

10. The power regulation system for energy-consuming equipment as described in claim 7, characterized in that, After generating the adjustment command based on the optimal power reduction amount, the method further includes: Obtain the actual power reduction amount after the pre-adjusted energy consumption device adjusts its power according to the optimal power reduction amount; Based on the actual power reduction amount and the optimal power reduction amount, the power adjustment deviation is obtained, and structured prompt words are generated according to the power adjustment deviation; Based on the structured prompts, the retrieval enhancement generation capability is analyzed under a preset large language model, and the device adjustment weight sequence is updated.

Images