UPS (Uninterrupted Power Supply) energy saving method based on load adaptive regulation and control

Abstract

The invention discloses a UPS (Uninterrupted Power Supply) energy-saving method based on load self-adaptive regulation and control, which relates to the technical field of UPS energy saving and load regulation and control, and comprises the following steps of: firstly, acquiring information such as current, voltage and a magnetic field through a high-precision multi-frequency-band sensor at high frequency, calculating load power by using a data fusion algorithm, and recording a change trend; and then identifying the load type by Fourier, wavelet transform and fuzzy logic, determining an energy-saving strategy according to the real-time condition and type of the load and magnetic field characteristics, regulating and controlling multiple parameters of the inverter, finally executing the strategy, and realizing accurate energy saving according to multi-index deviation feedback adjustment. The method is remarkable in energy saving, can accurately monitor the load, formulates a personalized strategy according to the real-time and type conditions of the load, reduces the energy consumption, enhances the adaptability to different loads, improves the reliability and stability of a power supply, has expansibility and compatibility, and provides intelligent and convenient experience for a user.

Description

Technical Field

[0001] The present invention relates to the technical field of energy saving and load regulation of UPS power supplies, and particularly to an energy saving method for UPS power supplies based on load adaptive regulation. Background Art

[0002] In today's digital age, uninterruptible power supplies (UPS) are key devices for ensuring stable power supply and are widely used in many fields such as data centers, industrial automation, and medical equipment. However, with the increasingly prominent global energy problem, the energy consumption problem of UPS power supplies has gradually become the focus of attention. During the operation of traditional UPS power supplies, regardless of how the load changes, they work in a relatively fixed mode, which results in serious energy waste under light load conditions. For example, in some data centers, the load is relatively low during nighttime or holiday periods, but the UPS power supply still maintains a high energy consumption level, causing a large amount of unnecessary power loss.

[0003] Existing energy saving technologies for UPS power supplies mainly focus on improving the conversion efficiency of the power supply itself, such as using new power devices and optimizing the circuit topology. However, these methods have poor adaptability to load changes and cannot adjust the working mode in real time according to the actual load conditions. Moreover, in a complex industrial environment, there are various types of loads, including linear loads, non-linear loads, and loads with periodic changes. Different types of loads have significant differences in the energy consumption impact on UPS power supplies. Traditional energy saving technologies are difficult to accurately identify different load types and perform targeted energy saving control, resulting in limited energy saving effects.

[0004] In addition, with the continuous development of technology, some advanced devices have higher and higher requirements for the stability and reliability of the power supply. During the energy saving process of existing UPS power supplies, due to inaccurate monitoring and regulation of the load, the power supply output may be unstable, affecting the normal operation of the device. For example, in some high-precision experimental devices, even a slight fluctuation in the power supply may cause deviations in the experimental results. Therefore, it is of great practical significance to develop an energy saving method for UPS power supplies that can adaptively regulate according to the load in real time. Summary of the Invention

[0005] The energy saving method for UPS power supplies based on load adaptive regulation proposed by the present invention aims to solve the problems mentioned in the above existing technologies.

[0006] To achieve the above object, the present invention adopts the following technical solutions: An energy saving method for UPS power supplies based on load adaptive regulation, comprising the following steps: Steps for real-time load monitoring: Using current and voltage sensors, the current I and voltage V at the output end of the UPS power supply are collected in real time. A magnetic field sensor is introduced to monitor the magnetic field changes around the load. By analyzing the correlation between the magnetic field intensity B and the load current, the load operating state is judged. According to the formula the load power is calculated, where is the power factor. In addition, the data fusion algorithm is used to fuse and process the data, record the load change trend, and judge the load dynamic characteristics.

[0007] Steps for load type identification: Wavelet transform is introduced to analyze the waveform and extract information. Combining with the fuzzy logic algorithm, the load type is judged. For loads in complex industrial environments, a load feature database is established, and the real-time collected features are matched with the typical features in the database to determine the specific load type.

[0008] Steps for formulating energy-saving strategies: According to the real-time power, type, change trend, and magnetic field characteristics of the load, an energy-saving strategy is formulated. In the case of light load, the deep sleep mode is adopted, and in the case of heavy load, the adaptive intelligent control algorithm is used to optimize the efficiency curve of the UPS power supply. According to the formula the switching frequency is adjusted, where f 0 is the initial switching frequency, and k 1 , k 2 are the adjustment coefficients, P rated is the rated power of the UPS power supply, the modulation ratio M is adjusted in real time according to the dynamic response characteristics of the load, and the phase angle is optimized according to the load power factor and magnetic field characteristics; Steps for strategy execution and feedback: The energy-saving strategy is sent to the control system of the UPS power supply, and the control system adjusts the working parameters of the UPS power supply according to the strategy. At the same time, the effect after the execution of the energy-saving strategy is monitored in real time. If the energy-saving effect does not meet the expectation, the energy-saving strategy is readjusted according to the multi-index deviation where is the power deviation, is the magnetic field intensity deviation, is the power factor deviation.

[0009] Furthermore, it also includes the load prediction step. Based on the data of historical load, magnetic field changes, and environmental temperature, a long short-term memory network deep learning model is used to predict the future load power. The formula for predicting the load power at the future h moment is , where P t , B t , T t are the current load power, magnetic field intensity, and environmental temperature respectively. According to the prediction results, the working state of the UPS power supply is adjusted in advance.

[0010] Further, it also includes a battery management step that monitors the battery voltage U in real time b , current I b , temperature T b , battery internal resistance R b , and uses the formula to calculate the charging current I charge , where I 0 is the initial charging current, SOC is the state of charge of the battery, R 0 is the initial internal resistance, is the change in magnetic field strength, k 3 , k 4 are adjustment coefficients. When the battery temperature is too high or the internal resistance increases abnormally, the charging power is reduced, and the battery's heat dissipation strategy is adjusted according to the magnetic field situation.

[0011] Further, in the load real-time monitoring step, the vibration information of the load is also collected, and the vibration frequency f v and amplitude A v of the load are obtained in real time through a vibration sensor, and the relationship between vibration and the load operation state is analyzed by combining the load power, magnetic field, and temperature. When the load vibration is abnormal, an alarm is issued and the energy-saving strategy is adjusted, and the vibration information is used to predict the faults of the load. By establishing an association model between vibration characteristics and fault types, potential fault hazards can be detected in a timely manner.

[0012] Further, in the load type identification step, a deep convolutional neural network is used to classify the load type. The collected multi-dimensional features are used as inputs, and after multiple convolutional and pooling operations of the DCNN model, the feature information is extracted and the load type is output; at the same time, transfer learning technology is introduced to adapt to new load types using existing similar load types.

[0013] Further, in the energy-saving strategy formulation step, for different types of loads, a multi-dimensional energy-saving strategy combination is adopted. For inductive loads, the power factor is improved by adjusting the capacitance value C of the capacitor compensation device, and the capacitance value is calculated according to the formula , where P is the load power, is the power supply angular frequency, U is the power supply voltage, is the power factor angle before compensation, is the power factor angle after compensation. For capacitive loads, a strategy combining inductive compensation and inverter parameter optimization is adopted. For equipment with periodic load characteristics, the output power and working mode of the UPS power supply are adjusted in advance according to the load cycle.

[0014] Further, in the strategy execution and feedback step, an energy-saving effect evaluation and traceability system based on blockchain technology is established, and the data is encrypted and stored on the blockchain. At the same time, an energy-saving effect evaluation index system is established. In addition to the energy-saving rate In addition to the power factor improvement rate, indicators such as the magnetic field energy efficiency ratio and the vibration energy consumption ratio are introduced to regularly evaluate the energy-saving effect, optimize the energy-saving strategy, and record the optimization information on the blockchain.

[0015] Furthermore, the method also includes communication and remote monitoring steps. The operation data of the UPS power supply is transmitted to the remote monitoring center through 5G communication technology. The monitoring center uses big data analysis and artificial intelligence algorithms to mine the data, generate equipment health reports and energy-saving suggestions. At the same time, VR and AR technologies provide equipment monitoring and operation experiences for maintenance personnel to achieve remote repair guidance. Through smart contract technology, automated transactions and settlements for the energy-saving services of the UPS power supply are realized.

[0016] Furthermore, in the load real-time monitoring step, first, the Kalman filter algorithm is used to preliminarily filter the data to remove noise and interference, and then the adaptive median filter algorithm is used for secondary processing of the data. The filtering formula is calculated according to the state prediction and update equations of the Kalman filter and the threshold judgment rules of the adaptive median filter.

[0017] Furthermore, in the energy-saving strategy formulation step, considering the full life cycle cost and environmental impact of the UPS power supply, for the UPS power supply using green and environmentally friendly materials and energy-saving technologies, the adjustment range of the energy-saving strategy is increased. According to the energy policies and electricity price fluctuations in different regions, the energy-saving strategy is adjusted. And an adaptive optimization mechanism for the energy-saving strategy is established to automatically adjust the strategy parameters and rules according to the actual operation of the UPS power supply.

[0018] Compared with the existing technologies, the beneficial effects of the present invention are as follows: By real-time monitoring multi-dimensional information such as the current, voltage, magnetic field, temperature, and vibration of the load, the load power can be accurately calculated, and personalized energy-saving strategies can be formulated according to the real-time changes and types of the load. In the light load condition, the deep sleep mode is adopted and the internal circuit of the module is optimized to reduce the standby power consumption; in the heavy load condition, the inverter parameters are optimized through the adaptive intelligent control algorithm to improve the conversion efficiency. This precise control method greatly reduces energy waste and operating costs.

[0019] This method improves the adaptability of the UPS power supply to different load types. Using deep learning models and multi-dimensional feature analysis, various types of loads such as linear loads and non-linear loads can be accurately identified, and multi-dimensional energy-saving strategy combinations are adopted for different types of loads. For example, for inductive loads, the power factor is improved by adjusting the capacitor compensation device, and for capacitive loads, a strategy combining inductance compensation and inverter parameter optimization is adopted, effectively improving the energy-saving effect and the stability of the power supply.

[0020] The method of this patent enhances the reliability and stability of the UPS power supply. By introducing technologies such as battery internal resistance monitoring and vibration monitoring, it can real-time grasp the battery status and load operation status, discover potential fault hazards in advance and adjust strategies in a timely manner. At the same time, it adopts multi-stage filtering processing and data fusion algorithms to ensure the accuracy and reliability of the collected data, providing strong support for the formulation and implementation of energy-saving strategies.

[0021] In addition, the method of this patent also has good scalability and compatibility. It realizes remote monitoring and big data analysis through 5G communication technology, uses blockchain technology to ensure the traceability of energy-saving data, and combines VR, AR technologies and smart contract technologies to provide users with a more intelligent and convenient service experience, adapting to the development trend of future energy management. Brief Description of the Drawings

[0022] Figure 1 It is a schematic block diagram of the energy-saving method of the UPS power supply based on load adaptive regulation proposed by the present invention; Figure 2 It is a bar chart of the energy-saving rate comparison of the energy-saving method of the UPS power supply based on load adaptive regulation proposed by the present invention; Figure 3 It is a diagram of the load power and energy-saving effect of the energy-saving method of the UPS power supply based on load adaptive regulation proposed by the present invention; Figure 4 It is a bar chart of the power factor improvement rate under different load types of the energy-saving method of the UPS power supply based on load adaptive regulation proposed by the present invention; Figure 5 It is a radar chart of the battery life comparison of the energy-saving method of the UPS power supply based on load adaptive regulation proposed by the present invention. Detailed Embodiment

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0025] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, the meaning of "a plurality" is two or more, unless otherwise specifically defined. In addition, the terms "mounted", "connected", and "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below with reference to the drawings.

[0026] Refer to Figures 1 to 5 : A UPS power supply energy-saving method based on load adaptive regulation, comprising the following steps: Load real-time monitoring step: Install high-precision, multi-band current sensors, voltage sensors, and magnetic field sensors at the output end of the UPS power supply. The current sensor and the voltage sensor collect the current I and the voltage V at a frequency of 20 times per second, and the magnetic field sensor synchronously monitors the magnetic field intensity B. At the same time, install temperature sensors and vibration sensors on the surface of the load device to collect the temperature T and vibration frequency f v , amplitude A v . The collected data is first preliminarily filtered by the Kalman filter algorithm to remove noise and interference. The state prediction equation of the Kalman filter is , and the update equation is , where is the state estimate value, F is the state transition matrix, B is the control input matrix, u is the control input, z is the measurement value, H is the measurement matrix, and K is the Kalman gain. Then, the adaptive median filtering algorithm is used to perform secondary processing on the filtered data to further eliminate impulse noise and outliers. After the filtering process, the current, voltage, magnetic field, temperature, and vibration data are fused using a data fusion algorithm, and the load power P is calculated according to the formula where is dynamically obtained through a real-time power factor analyzer. Meanwhile, the rising or falling rate of the load power and the rate of change of the magnetic field intensity are recorded to provide a basis for subsequent load type identification and energy-saving strategy formulation.

[0027] Load type identification steps: First, the collected load current and voltage waveforms are processed. Using Fourier transform, the waveforms are transformed into the frequency domain to accurately analyze the harmonic components therein. At the same time, wavelet transform is introduced to deeply analyze the waveforms from multiple scales and extract more abundant characteristic information. Then, combined with the fuzzy logic algorithm, comprehensively considering waveform characteristics, harmonic content, magnetic field changes, temperature, and vibration information, a comprehensive and accurate comprehensive judgment of the load type is made. Moreover, in order to further improve the accuracy of identification, a load characteristic database needs to be established, and the real-time collected characteristics are matched with the typical characteristics in the database. Especially for loads in complex industrial environments, such as high-precision CNC machine tools, automated production line equipment, etc., the characteristic information in the database needs to be continuously updated to improve the accuracy of load type identification and better meet the requirements of complex and changing industrial scenarios.

[0028] Energy-saving strategy formulation steps: According to the real-time power, type, change trend, and magnetic field characteristics of the load, corresponding energy-saving strategies are formulated. For light loads (load power less than 15% of the rated power of the UPS power supply), a deep sleep mode is adopted, not only turning off some redundant UPS modules, but also optimizing and adjusting the circuits inside the modules to reduce standby power consumption. For heavy loads (load power greater than 85% of the rated power of the UPS power supply), an adaptive intelligent control algorithm is used to optimize the working efficiency curve of the UPS power supply. By adjusting the switching frequency f s , modulation ratio M, and phase angle to improve the conversion efficiency. The specific switching frequency f s is adjusted according to the formula where f 0 is the initial switching frequency, k 1 , k 2 are adjustment coefficients, and P rated is the rated power of the UPS power supply. The modulation ratio M is adjusted in real time according to the dynamic response characteristics of the load, and the phase angle Optimize according to the power factor and magnetic field characteristics of the load.

[0029] Policy execution and feedback step: Send the formulated energy-saving policy to the control system of the UPS power supply. The control system precisely adjusts the working parameters of the UPS power supply according to the policy. At the same time, monitor the effect after the execution of the energy-saving policy in real time, and compare multiple indicators such as load power, energy consumption, magnetic field strength, power factor, temperature, and vibration before and after the execution. If the energy-saving effect does not meet the expectation, according to the multi-finger deviation Readjust the energy-saving policy, where is the power deviation, is the magnetic field strength deviation, is the power factor deviation.

[0030] In the present invention, a load prediction step is further included. This step is based on historical load data, magnetic field change data, and environmental temperature data, and uses a long short-term memory network (LSTM) deep learning model to predict the load power in a future period. The prediction formula combines time series information and multi-variable features, such as , where is the predicted load power at the future h-th moment, P t is the load power at the current moment, B t is the magnetic field strength at the current moment, T t is the environmental temperature at the current moment. Adjust the working state of the UPS power supply in advance according to the prediction result, and optimize the adjustment strategy considering the magnetic field change trend to avoid frequent state switching.

[0031] In the present invention, a battery management step is further included. This step monitors the state of the UPS power supply battery in real time. In addition to monitoring parameters such as the battery voltage U b , current I b , and temperature T b , a battery internal resistance monitoring module is introduced to measure the battery internal resistance R b in real time. Adjust the charging current and charging time according to the battery charge and discharge curve, health status, internal resistance change, and the influence of the magnetic field on the battery. Use the formula to calculate the charging current I charge , where I 0 is the initial charging current, SOC is the state of charge of the battery, R 0 is the initial internal resistance of the battery, is the change in the magnetic field strength around the battery, k 3 , k 4 are adjustment coefficients. When the battery temperature is too high or the internal resistance increases abnormally, reduce the charging power, and at the same time adjust the battery heat dissipation strategy according to the magnetic field situation to protect the battery life.

[0032] In the present invention, in the step of real-time monitoring of the load, in order to comprehensively and accurately grasp the operating state of the load, in addition to collecting current, voltage, magnetic field, and temperature information, the collection of load vibration information is also added. With the help of a high-precision vibration sensor, the vibration frequency f of the load is obtained in real time. v and the amplitude A v , and spectrum analysis and trend monitoring are performed on the collected vibration data. In the data processing stage, the vibration information is comprehensively analyzed together with the load power, magnetic field, and temperature data to deeply explore the relationship between vibration and the operating state of the load. Once the system determines that the load vibration is abnormal, an alarm will be immediately issued, and according to the preset rules, the energy-saving strategy will be intelligently adjusted to avoid equipment damage or increased energy consumption caused by vibration. At the same time, using the vibration data accumulated over a long time, a correlation model between vibration characteristics and fault types based on machine learning algorithms is built to conduct early warning of the faults of the load. By real-time monitoring of the vibration data and comparing it with the model in a timely manner, potential fault hazards can be quickly discovered, leaving sufficient time for maintenance personnel to conduct fault troubleshooting and repair, thereby reducing the economic losses caused by equipment failures and improving the stability and reliability of the system.

[0033] In the present invention, in the step of load type identification, a deep convolutional neural network (DCNN) is innovatively used to classify the load types. Multidimensional features such as current, voltage, magnetic field, temperature, and vibration during the operation of the load are collected as the input of the DCNN model. After these data enter the model, they will go through multiple layers of convolution and pooling operations to extract deep-level feature information, and then the load type is output. To continuously improve the performance of the model, by continuously updating the training data and incorporating the load characteristics under different working conditions, the classification accuracy and generalization ability are greatly enhanced. At the same time, transfer learning technology is introduced, and with the help of existing classification models for similar load types, the new load type identification task can be quickly adapted, reducing the model training cost and shortening the identification cycle, helping the load classification in industrial scenarios to be more efficient and accurate.

[0034] In the present invention, in the step of formulating the energy-saving strategy, for different types of loads, a multi-dimensional energy-saving strategy combination is adopted. For inductive loads, in addition to adjusting the inverter efficiency and using harmonic suppression devices, the power factor is improved by adjusting the capacitance value C of the capacitor compensation device, and the capacitance value is calculated according to the formula , where P is the load power, is the power supply angular frequency, U is the power supply voltage, is the power factor angle before compensation, is the power factor angle after compensation. For capacitive loads, a strategy combining inductive compensation and inverter parameter optimization is adopted. For equipment with periodic load characteristics, the output power and working mode of the UPS power supply are adjusted in advance according to the load cycle to achieve precise energy saving.

[0035] In the present invention, in the strategy execution and feedback step, an energy-saving effect evaluation and traceability system based on blockchain technology is established. Various data (including load power, energy consumption, magnetic field intensity, power factor, etc.) during the execution of the energy-saving strategy are encrypted and stored on the blockchain to ensure the immutability and traceability of the data. At the same time, an energy-saving effect evaluation index system is established. In addition to the energy-saving rate , in addition to the power factor improvement rate, indexes such as magnetic field energy efficiency ratio and vibration energy consumption ratio are also introduced. The energy-saving effect is evaluated regularly, the energy-saving strategy is optimized according to the evaluation results, and the optimization information is recorded on the blockchain to provide a reliable basis for subsequent energy-saving improvements.

[0036] In the present invention, the method further includes a communication and remote monitoring step. Through the high bandwidth and low latency characteristics of 5G communication technology, multiple data such as the load power, energy consumption, and battery status of the UPS power supply operation are transmitted to the remote monitoring center at high speed and stably. In the monitoring center, massive data is deeply mined with the help of big data analysis and artificial intelligence algorithms to generate accurate equipment health reports and targeted energy-saving suggestions in real time. At the same time, virtual reality (VR) and augmented reality (AR) technologies are used to create an immersive equipment monitoring and operation scenario for maintenance personnel to assist in remote fault diagnosis and repair guidance. In addition, through smart contract technology, an automated transaction and settlement system is built to ensure the efficient and safe development of UPS power supply energy-saving services in all links, realizing full-spectrum intelligence from data collection to commercial application.

[0037] In the present invention, in the load real-time monitoring step, strict multi-level filtering processing is performed on the collected data. First, the Kalman filtering algorithm is used. According to its unique state prediction and update equations, preliminary filtering operations are carried out on data such as current, voltage, and magnetic field, which can effectively remove the noise and interference generated during the circuit transmission process and significantly improve the accuracy of the data. Immediately afterwards, the adaptive median filtering algorithm is used to perform secondary processing on the data after Kalman filtering. By setting reasonable threshold judgment rules, impulse noise and outliers caused by abnormal fluctuations of the equipment are further eliminated. Through these two-step filtering, the collected data can truly, stably, and reliably reflect the operating state of the load, providing solid data support for subsequent load regulation, energy-saving strategy formulation, etc., and strongly guaranteeing the stability and accuracy of the system operation.

[0038] In the present invention, in the step of formulating the energy-saving strategy, when evaluating the full life-cycle cost of the UPS power supply, not only the initial procurement cost is analyzed, but also the long-term operation and maintenance cost and the end-of-life treatment cost are included; when considering the environmental impact, attention is paid to its carbon emissions and resource consumption during the use stage. For UPS power supplies using green and environmentally friendly materials and energy-saving technologies, with the help of an intelligent control system, the adjustment range of the energy-saving strategy is appropriately increased to give full play to their energy-saving advantages. At the same time, a real-time data monitoring platform is built, and with the help of big data analysis, the energy policies and electricity price fluctuation rules in different regions are closely tracked to achieve dynamic adjustment of the energy-saving strategy. For example, during the peak electricity price period, more aggressive energy-saving measures such as optimizing the inverter operation mode and reducing the redundant power consumption of equipment are taken to reduce the electricity cost; during the low electricity price period, the energy-saving requirements are appropriately relaxed to meet the needs of efficient operation of the equipment, so as to maximize the economic and environmental benefits. In addition, an adaptive optimization mechanism for the energy-saving strategy is established. According to the actual operation status and energy-saving effect feedback of the UPS power supply, intelligent algorithms are used to automatically adjust the strategy parameters and rules, continuously improve the energy-saving efficiency, and ensure the sustainable operation of the UPS power supply throughout its life cycle. During the implementation of the strategy, blockchain technology is used to record each strategy adjustment and implementation result to ensure the immutability and traceability of the data, providing a reliable basis for the optimization and iteration of the strategy.

[0039] The above is only the preferred specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

Claims

1. A UPS power energy saving method based on load adaptive control, characterized in that: The following steps are involved: Load real-time monitoring steps: Use current and voltage sensors to collect the current I and voltage V at the output end of the UPS power supply in real time, introduce magnetic field sensors to monitor the magnetic field changes around the load, and judge the load operation status by analyzing the relationship between the magnetic field strength B and the load current. Calculate the load power, where In addition, the data fusion algorithm is used to fuse the data, record the load change trend, and judge the dynamic characteristics of the load; Load type identification steps: Introduce wavelet transform to analyze the waveform, extract information, and combine fuzzy logic algorithm to judge the load type. For loads in complex industrial environments, establish a load feature database, match the real-time collected features with the typical features in the database, and determine the specific type of the load; Energy-saving strategy formulation steps: formulate energy-saving strategies based on the real-time power, type, change trend and magnetic field characteristics of the load. In light-load conditions, the deep sleep mode is adopted. In heavy-load conditions, the adaptive intelligent control algorithm is used to optimize the UPS power efficiency curve. According to the formula Adjust the switching frequency, f0 is the initial switching frequency, k1 and k2 are the adjustment coefficients, P rated is the rated power of the UPS power supply, the modulation ratio M is adjusted in real time according to the dynamic response characteristics of the load, and the phase angle Optimize according to load power factor and magnetic field characteristics; Strategy execution and feedback steps: The energy-saving strategy is sent to the UPS power supply control system. The control system adjusts the UPS power supply operating parameters according to the strategy. At the same time, the effect of the energy-saving strategy execution is monitored in real time. If the energy-saving effect does not meet expectations, it will be monitored based on the multi-indicator deviation. Re-adjust energy-saving strategies, including is the power deviation, is the magnetic field strength deviation, is the power factor deviation.

2. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: It also includes a load prediction step, which uses a long short-term memory network deep learning model to predict future load power based on historical load, magnetic field change, and ambient temperature data. The formula for predicting load power at the next h moment is: , P t , B t 、T t They are the current load power, magnetic field strength, and ambient temperature respectively, and the UPS power supply working state is adjusted in advance according to the prediction results.

3. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: It also includes a battery management step, which monitors the battery voltage U in real time. b 、Current I b , Temperature T b 、Battery internal resistance R b , using the formula Calculate the charging current I charge , where I0 is the initial charging current, SOC is the battery state of charge, and R0 is the initial internal resistance. is the change of magnetic field strength, k3 and k4 are adjustment coefficients. When the battery temperature is too high or the internal resistance increases abnormally, the charging power is reduced and the heat dissipation strategy of the battery is adjusted according to the magnetic field conditions.

4. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: In the load real-time monitoring step, the vibration information of the load is also collected, and the vibration frequency f of the load is obtained in real time through the vibration sensor. v and amplitude A v , and combines load power, magnetic field, and temperature to analyze the relationship between vibration and load operating status. When the load vibration is abnormal, an alarm is issued and the energy-saving strategy is adjusted. The vibration information is used to warn of load failures. By establishing a correlation model between vibration characteristics and fault types, potential fault hazards can be discovered in a timely manner.

5. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: In the load type identification step, a deep convolutional neural network is used to classify the load type. The collected multi-dimensional features are used as input. After multi-layer convolution and pooling operations of the DCNN model, feature information is extracted and the load type is output. At the same time, transfer learning technology is introduced to use existing similar load types to adapt to new load types.

6. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: In the energy-saving strategy formulation step, for different types of loads, a multi-dimensional energy-saving strategy combination is adopted. For inductive loads, the power factor is improved by adjusting the capacitance value C of the capacitor compensation device. The capacitance value is calculated according to the formula Calculate, where P is the load power, is the power supply angular frequency, U is the power supply voltage, is the power factor angle before compensation, To obtain the power factor angle after compensation, for capacitive loads, a strategy combining inductance compensation and inverter parameter optimization is adopted. For equipment with periodic load characteristics, the output power and working mode of the UPS power supply are adjusted in advance according to the load cycle.

7. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: In the strategy execution and feedback steps, an energy-saving effect evaluation and traceability system based on blockchain technology is established, and data is encrypted and stored on the blockchain. At the same time, an energy-saving effect evaluation index system is established. In addition to the energy-saving rate In addition to the power factor improvement rate, the magnetic field energy efficiency ratio and vibration energy consumption ratio indicators are also introduced to regularly evaluate the energy-saving effect, optimize the energy-saving strategy, and record the optimization information on the blockchain.

8. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: The method also includes communication and remote monitoring steps. The operating data of the UPS power supply is transmitted to the remote monitoring center through 5G communication technology. The monitoring center uses big data analysis and artificial intelligence algorithms to mine data and generate equipment health reports and energy-saving suggestions. At the same time, VR and AR technologies provide equipment monitoring and operation experience for operation and maintenance personnel, realize remote repair guidance, and realize automated transactions and settlements of UPS power supply energy-saving services through smart contract technology.

9. The UPS power source energy saving method based on load adaptive control according to claim 1 is characterized in that: In the real-time load monitoring step, the Kalman filter algorithm is first used to perform preliminary filtering on the data to remove noise and interference, and then the adaptive median filter algorithm is used to perform secondary processing on the data. The filtering formula is calculated based on the state prediction and update equation of the Kalman filter and the threshold judgment rule of the adaptive median filter.

10. The UPS power source energy saving method based on load adaptive control according to claim 1, characterized in that: In the energy-saving strategy formulation step, the full life cycle cost and environmental impact of the UPS power supply are considered. For UPS power supplies that use green and environmentally friendly materials and energy-saving technologies, the adjustment range of the energy-saving strategy is increased. The energy-saving strategy is adjusted according to the energy policies and electricity price fluctuations in different regions. In addition, an adaptive optimization mechanism for the energy-saving strategy is established to automatically adjust the strategy parameters and rules according to the actual operation of the UPS power supply.

Images