Vacuum pump energy consumption real-time monitoring and analyzing method and system

Abstract

The invention relates to the technical field of intelligent manufacturing, and discloses a vacuum pump energy consumption real-time monitoring and analysis method and system, and the method comprises the steps: carrying out the multi-scale filtering of a current signal and a voltage signal of a load loop in a vacuum pump, and obtaining a current waveform and a voltage waveform of the load loop; performing time-frequency domain feature analysis on the current waveform, and performing time domain behavior analysis on the voltage waveform to obtain a current feature set and a voltage feature set of a load loop; performing association fusion on the current feature set and the voltage feature set to obtain an electric energy quality feature sequence of the load loop; based on the electric energy quality characteristic sequence, carrying out characteristic adaptability correction on the current waveform and the voltage waveform to obtain a current effective value and a voltage effective value of a load loop; the current effective value and the voltage effective value are comprehensively integrated to obtain an accurate electric energy measurement value of the vacuum pump; practicality and guidance of real-time monitoring and analysis of energy consumption of the vacuum pump can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent manufacturing, and particularly relates to a vacuum pump energy consumption real-time monitoring and analysis method and system. BACKGROUND

[0002] In the field of vacuum pump energy consumption real-time monitoring and analysis, the existing technology has obvious defects in the processing link of load loop current signal and voltage signal. Most of the existing methods use single-scale filtering means, which cannot comprehensively eliminate the interference components in the signal, for example, it is difficult to effectively separate the power frequency noise, suppress high frequency oscillation and correct baseline drift, and at the same time, the transient pulse and common mode noise in the voltage signal are not thoroughly processed, resulting in a large distortion of the current waveform and voltage waveform obtained finally. Such distortion will directly affect the accuracy of subsequent current and voltage feature extraction, making it impossible to obtain feature information that can truly reflect the running state of the load loop, and hiding precision hazards for subsequent energy consumption calculation.

[0003] The existing technology also has deficiencies in the integration link of feature processing and electric energy measurement. On the one hand, the existing method lacks effective correlation fusion mechanism for the extracted current feature set and voltage feature set, and does not perform accurate time sequence alignment and cross-correlation analysis, resulting in the inability to construct an electric energy quality feature sequence that can accurately reflect the internal relationship between the two; on the other hand, when performing waveform correction based on features, there is a lack of scientific matching and comparison with reference features, making it difficult to accurately identify feature deviation patterns, and thus unable to accurately adaptively correct the current waveform and voltage waveform, resulting in a large error in the calculated current effective value and voltage effective value. At the same time, the existing technology does not perform strict integrity verification and historical reference comparison analysis when integrating effective value data, and the final electric energy measurement value cannot accurately reflect the actual energy consumption of the vacuum pump, making it difficult to meet the needs of real-time monitoring and accurate analysis. SUMMARY

[0004] The present application provides a vacuum pump energy consumption real-time monitoring and analysis method and system to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides a vacuum pump energy consumption real-time monitoring and analysis method, comprising: S1. Multi-scale filtering the current signal and voltage signal of the load loop in the vacuum pump to obtain the current waveform and voltage waveform of the load loop; S2. Time-frequency domain feature analysis on the current waveform and time domain behavior analysis on the voltage waveform to obtain the current feature set and voltage feature set of the load loop; S3. Correlation fusion of the current feature set and the voltage feature set to obtain the electric energy quality feature sequence of the load loop; S4. Based on the sequence of power quality characteristics, the current waveform and the voltage waveform are adaptively corrected, and the current effective value and the voltage effective value of the load circuit are obtained; S5. The current effective value and the voltage effective value are integrated, and the accurate power measurement value of the vacuum pump is obtained.

[0006] In a preferred embodiment, the current signal and the voltage signal of the load circuit in the vacuum pump are multi-scale filtered to obtain the current waveform and the voltage waveform of the load circuit, including: The power frequency noise of the current signal is separated to obtain the first scale filtered current signal of the load circuit; The high-frequency oscillation of the first scale filtered current signal is suppressed to obtain the second scale filtered current signal of the load circuit; The baseline drift of the second scale filtered current signal is corrected to obtain the current waveform of the load circuit; The transient pulse of the voltage signal is filtered to obtain the pretreated voltage signal of the load circuit; The common mode noise of the pretreated voltage signal is suppressed to obtain the voltage waveform of the load circuit.

[0007] In a preferred embodiment, the current waveform is analyzed in time and frequency domain, and the voltage waveform is analyzed in time domain to obtain the current characteristic set and the voltage characteristic set of the load circuit, including: The frequency domain characteristics of the current waveform are extracted to obtain the main frequency component distribution characteristics and the spectral energy concentration characteristics, and the main frequency component distribution characteristics and the spectral energy concentration characteristics are taken as the current frequency domain characteristic subset of the load circuit; The time-varying characteristics of the current waveform are analyzed to obtain the amplitude envelope characteristics and the waveform distortion evolution characteristics, and the amplitude envelope characteristics and the waveform distortion evolution characteristics are taken as the current time domain characteristic subset of the load circuit; The current frequency domain characteristic subset and the current time domain characteristic subset are integrated to obtain the current characteristic set of the load circuit; The steady state characteristics of the voltage waveform are evaluated to obtain the amplitude stability characteristics and the waveform symmetry characteristics, and the amplitude stability characteristics and the waveform symmetry characteristics are taken as the voltage steady state characteristic subset of the load circuit; The dynamic process of the voltage waveform is tracked to obtain the transient response characteristics and the recovery characteristics, and the transient response characteristics and the recovery characteristics are taken as the voltage dynamic characteristic subset of the load circuit; merge the voltage steady-state feature subset and the voltage dynamic feature subset to obtain a voltage feature set of the load circuit.

[0008] In a preferred embodiment, the associating and fusing the current feature set and the voltage feature set to obtain the power quality feature sequence of the load circuit comprises: performing time alignment processing on the current feature set and the voltage feature set to obtain a time-calibrated feature set of the load circuit; performing cross-correlation analysis on the time-calibrated feature set to obtain a feature correlation matrix of the load circuit; structuring and reconstructing the feature correlation matrix to obtain a current-voltage feature mapping relationship table of the load circuit; based on the time dimension set and the feature dimension set, serializing and recombining the current-voltage feature mapping relationship table to obtain the power quality feature sequence of the load circuit.

[0009] In a preferred embodiment, the performing cross-correlation analysis on the time-calibrated feature set to obtain the feature correlation matrix of the load circuit comprises: obtaining a cross-correlation coefficient of the current feature set and the voltage feature set in the time domain to obtain a time-domain correlation index of the load circuit; performing frequency domain characteristic analysis on the current feature set and the voltage feature set to obtain a frequency domain correlation degree index of the load circuit; associating and fusing the time-domain correlation index and the frequency domain correlation degree index to obtain the feature correlation matrix of the load circuit.

[0010] In a preferred embodiment, the serializing and recombining the current-voltage feature mapping relationship table based on the time dimension set and the feature dimension set to obtain the power quality feature sequence of the load circuit comprises: performing time sequence arrangement on feature data in the current-voltage feature mapping relationship table according to the acquisition time sequence to obtain a time dimension sequence of the load circuit; performing secondary sorting on the feature data according to the feature importance degree to obtain a feature dimension sequence of the load circuit; performing double indexing on the time dimension sequence and the feature dimension sequence to obtain the power quality feature sequence of the load circuit; In a preferred embodiment, the performing feature adaptability correction on the current waveform and the voltage waveform based on the power quality feature sequence to obtain the current effective value and the voltage effective value of the load circuit comprises: The power quality feature sequence is matched and compared with a preset benchmark feature library to obtain the feature deviation pattern of the load circuit. Based on the characteristic deviation mode, the current waveform and the voltage waveform are parametrically adjusted to obtain the current waveform correction parameters and voltage waveform correction parameters of the load circuit; Based on the current waveform correction parameters and the voltage waveform correction parameters, the current waveform and the voltage waveform are dynamically reconstructed to obtain the reconstructed current waveform and the reconstructed voltage waveform of the load circuit. The reconstructed current waveform and the reconstructed voltage waveform are converted to RMS values ​​to obtain the RMS values ​​of the current and voltage of the load circuit.

[0011] In a preferred embodiment, the step of matching and comparing the power quality feature sequence with a preset reference feature library to obtain the feature deviation pattern of the load circuit includes: Based on a pre-set benchmark feature library, the power quality feature sequence is evaluated for similarity to obtain a similarity metric for the load circuit. The calculation formula for the similarity metric is as follows: ; In the formula, The similarity metric value, The first in the power quality characteristic sequence One feature parameter, The first standard feature in the benchmark feature library One feature parameter, The average value of the characteristic parameters of the power quality characteristic sequence. The average value of the standard characteristic parameters. Indicates the total number of feature parameters; Based on the similarity measurement results, the deviation feature quantity of the power quality feature sequence relative to the benchmark feature template is extracted; The distribution characteristics of the deviation features are analyzed to identify the characteristic deviation patterns of the load circuit.

[0012] In a preferred embodiment, the step of integrating the effective values ​​of the current and voltage to obtain the accurate electrical energy measurement value of the load circuit includes: The effective values ​​of current and voltage are paired and combined to obtain the electrical parameter data set of the load circuit; The integrity of the electrical parameter data set is verified to obtain the set of electrical parameters that have passed the verification of the load circuit. Based on the preset data structure and storage format, the verified set of electrical parameters is standardized and packaged to obtain a standardized electrical parameter data set of the load circuit; The standardized electrical parameter data set is compared and analyzed with historical electrical parameter benchmarks to obtain a characteristic mode and a change trend of the standardized electrical parameter data; The characteristic mode and the change trend are integrated to obtain an electric energy quality report of the load circuit; The electric energy quality report is used as an accurate electric energy measurement value of the vacuum pump.

[0013] To solve the above problems, the present application also provides a vacuum pump energy consumption real-time monitoring and analysis system, which comprises: A multi-domain feature analysis module is used to analyze the time-frequency domain features of the current waveform and the time-domain behavior of the voltage waveform to obtain a current feature set and a voltage feature set of the load circuit; A feature correlation and sequence fusion module is used to correlate and fuse the current feature set and the voltage feature set to obtain an electric energy quality feature sequence of the load circuit; A feature adaptive correction module is used to correct the current waveform and the voltage waveform based on the electric energy quality feature sequence to obtain a current effective value and a voltage effective value of the load circuit; An electric energy integration and output module is used to integrate the current effective value and the voltage effective value to obtain an accurate electric energy measurement value of the load circuit.

[0014] Compared with the prior art, the present application has the following beneficial effects: 1. The present technology has a significant improvement effect in the electric energy signal processing and feature extraction link of the vacuum pump load circuit. It can obtain undistorted current waveform and voltage waveform by sequentially completing power frequency noise separation, high frequency oscillation suppression and baseline drift correction on the current signal through multi-scale filtering, and by completing transient pulse filtering and common mode noise suppression on the voltage signal. At the same time, the main frequency component distribution, frequency domain features such as spectral energy concentration, and time domain features such as amplitude envelope and waveform distortion evolution are extracted from the current waveform through time-frequency domain feature analysis, and the amplitude stability, waveform symmetry and other steady-state features, and transient response, recovery characteristics and other dynamic features are extracted from the voltage waveform through time-domain behavior analysis, ensuring the comprehensiveness and accuracy of the current feature set and the voltage feature set, providing a high-quality data basis for subsequent electric energy quality analysis, and effectively improving the reliability and integrity of feature extraction.

[0015] 2.The technology has outstanding effect in the improvement of electric energy parameter correction and accurate measurement. Based on the matching comparison of power quality characteristic sequence and pre-set reference characteristic library, the characteristic deviation mode is accurately identified, and the current waveform and voltage waveform are parameterized adjusted and dynamically reconstructed, so that the current effective value and voltage effective value with high precision can be obtained. Then, through electric parameter pairing combination and integrity check, effective data screening is realized, combined with pre-set data structure to complete standardized packaging, and compared with historical electric parameter reference to obtain characteristic mode and change trend, and finally the generated power quality report is used as the accurate electric energy measurement value, which not only greatly improves the precision of vacuum pump energy consumption measurement, but also can present the energy consumption change law in real time, provides accurate data support for energy consumption monitoring and analysis, and significantly improves the practicability and guidance of energy consumption monitoring and analysis. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of a vacuum pump energy consumption real-time monitoring and analysis method provided by an embodiment of the present application is shown in the figure. Figure 2 A functional module diagram of a vacuum pump energy consumption real-time monitoring and analysis system provided by an embodiment of the present application is shown in the figure. The implementation, functional characteristics and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0017] It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0018] The embodiment of the present application provides a vacuum pump energy consumption real-time monitoring and analysis method. The execution subject of the vacuum pump energy consumption real-time monitoring and analysis method includes but is not limited to at least one of the electronic devices which can be configured to execute the method provided by the embodiment of the present application, such as a server and a terminal. In other words, the vacuum pump energy consumption real-time monitoring and analysis method can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be a stand-alone server, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and big data and artificial intelligence platforms, etc. Basic cloud computing services.

[0019] Referring to Figure 1 A flowchart of a vacuum pump energy consumption real-time monitoring and analysis method provided by an embodiment of the present application is shown in the figure. In this embodiment, the vacuum pump energy consumption real-time monitoring and analysis method includes: S1, multi-scale filtering is performed on the current signal and the voltage signal of the load circuit in the vacuum pump, to obtain a current waveform and a voltage waveform of the load circuit; In the embodiment of the present application, the multi-scale filtering performed on the current signal and the voltage signal of the load circuit in the vacuum pump to obtain the current waveform and the voltage waveform of the load circuit comprises: The power frequency noise separation is performed on the current signal to obtain a first scale filtered current signal of the load circuit; The high-frequency oscillation suppression is performed on the first scale filtered current signal to obtain a second scale filtered current signal of the load circuit; The baseline drift correction is performed on the second scale filtered current signal to obtain the current waveform of the load circuit; The transient pulse filtering is performed on the voltage signal to obtain a pretreated voltage signal of the load circuit; The common-mode noise suppression is performed on the pretreated voltage signal to obtain the voltage waveform of the load circuit.

[0020] Specifically, when the power frequency noise separation is performed on the current signal, a band-stop filter circuit is adopted, which is composed of inductance and capacitance with specific parameters, and the design target is to prevent the signal of power frequency from passing through. When the current signal enters the band-stop filter circuit, the noise signal of power frequency is significantly attenuated by the circuit, while the useful current signal of other frequencies can pass through normally. After the processing of the circuit, the output signal is the first scale filtered current signal of the load circuit.

[0021] Further, when the high-frequency oscillation suppression is performed on the first scale filtered current signal, a low-pass filter circuit is adopted, which is composed of resistance and capacitance, and has the characteristic of allowing signals below the set frequency to pass through, while the high-frequency oscillation signals above the set frequency are attenuated by the circuit. After the first scale filtered current signal flows through the low-pass filter circuit, the high-frequency oscillation components contained therein are effectively filtered out, and the remaining signal is the second scale filtered current signal of the load circuit.

[0022] Further, when the baseline drift correction is performed on the second scale filtered current signal, a correction circuit composed of a blocking direct current capacitor is adopted, which has the characteristic of preventing direct current signals from passing through and allowing alternating current signals to pass through. After the second scale filtered current signal passes through the blocking direct current capacitor, the direct current component contained therein that causes baseline drift is completely blocked, and only the alternating current signal is retained. The signal obtained after this processing is the current waveform of the load circuit.

[0023] Further, when the transient pulse of the voltage signal is filtered, a filter device composed of a transient voltage suppressor is adopted. The transient voltage suppressor is in a high resistance state under normal voltage, and when a transient pulse appears in the voltage signal and its amplitude exceeds a set threshold, it will quickly change to a low resistance state to release the energy of the transient pulse through the ground terminal. After the voltage signal is processed by the device, the transient pulse in the voltage signal is completely removed, and the output signal is the preprocessed voltage signal of the load circuit.

[0024] Further, when the common mode noise of the preprocessed voltage signal is suppressed, a suppression circuit composed of a common mode inductor is adopted. When the common mode noise passes through, the magnetic fields generated by the two windings of the common mode inductor are in the same direction and superimposed on each other, forming a larger impedance to prevent the common mode noise from passing through. When the useful differential mode voltage signal passes through, the magnetic fields generated by the two windings cancel each other out and can be transmitted smoothly. After the preprocessed voltage signal is processed by the circuit, the common mode noise is effectively suppressed, and the output signal is the voltage waveform of the load circuit.

[0025] In summary, the power frequency noise in the current signal is separated by removing the power frequency noise contained in the current signal through a specific filtering process, thereby obtaining the first scale filtered current signal of the load circuit.

[0026] In summary, the high-frequency oscillation in the first scale filtered current signal is suppressed by filtering the high-frequency oscillation components in the signal to eliminate such interference, thereby obtaining the second scale filtered current signal of the load circuit.

[0027] In summary, the baseline drift correction of the second scale filtered current signal is to keep the signal reference stable by processing and correcting the baseline offset phenomenon in the signal, and finally obtain the current waveform of the load circuit.

[0028] In summary, the transient pulse of the voltage signal is filtered by removing the transient pulse interference in the voltage signal to obtain a more stable signal, i.e., the preprocessed voltage signal of the load circuit.

[0029] In summary, the common mode noise of the preprocessed voltage signal is suppressed by processing and eliminating the common mode noise interference in the signal to ensure the purity of the signal, and finally obtain the voltage waveform of the load circuit.

[0030] S2, time-frequency domain feature analysis of the current waveform and time domain behavior analysis of the voltage waveform are performed to obtain the current feature set and voltage feature set of the load circuit. In the embodiment of the present application, the current waveform is analyzed in time-frequency domain, and the voltage waveform is analyzed in time domain, to obtain the current feature set and the voltage feature set of the load circuit, comprising: The current waveform is analyzed in frequency domain to obtain the main frequency component distribution feature and the spectral energy concentration degree feature of the current waveform, and the main frequency component distribution feature and the spectral energy concentration degree feature are taken as the current frequency domain feature subset of the load circuit. The current waveform is analyzed in time-varying feature to obtain the amplitude envelope feature and the waveform distortion evolution feature of the current waveform, and the amplitude envelope feature and the waveform distortion evolution feature are taken as the current time domain feature subset of the load circuit. The current frequency domain feature subset and the current time domain feature subset are integrated to obtain the current feature set of the load circuit. The voltage waveform is evaluated in steady state to obtain the amplitude stability feature and the waveform symmetry feature of the voltage waveform, and the amplitude stability feature and the waveform symmetry feature are taken as the voltage steady state feature subset of the load circuit. The voltage waveform is tracked in dynamic process to obtain the transient response feature and the recovery characteristic feature of the voltage waveform, and the transient response feature and the recovery characteristic feature are taken as the voltage dynamic feature subset of the load circuit. The voltage steady state feature subset and the voltage dynamic feature subset are combined to obtain the voltage feature set of the load circuit.

[0031] Specifically, when the current waveform is analyzed in frequency domain, a hardware spectrum analysis module is adopted. After receiving the current waveform, the module converts the time domain current waveform into a frequency domain signal through an internal signal conversion circuit, then identifies a number of frequency components with the highest energy proportion in the frequency domain signal, records the distribution state of these frequency components in the entire frequency range to form the main frequency component distribution feature of the current waveform; at the same time, the module calculates the proportion relationship between the sum of the energy of the main frequency components and the total energy to determine the concentration degree of the spectral energy, forms the spectral energy concentration degree feature, and integrates the main frequency component distribution feature and the spectral energy concentration degree feature to constitute the current frequency domain feature subset of the load circuit.

[0032] Furthermore, when performing time-varying characteristic analysis on the current waveform, an amplitude tracking circuit and a waveform comparison circuit are used. The amplitude tracking circuit continuously monitors the peak value change of the current waveform, records the peak value at each moment in real time, and connects them to form a smooth curve. This curve reflects the trend of the current amplitude change over time, which is the amplitude envelope characteristic of the current waveform. The waveform comparison circuit compares the actual current waveform with the standard sine waveform point by point, records the differences between the two in shape and phase, tracks the change of this difference over time, and forms a continuous change trajectory, which is the waveform distortion evolution characteristic. The amplitude envelope characteristic and the waveform distortion evolution characteristic are integrated together to form a subset of the current time-domain characteristics of the load circuit.

[0033] Furthermore, when integrating the current frequency domain feature subset and the current time domain feature subset, it is done through a data integration unit. This unit receives all feature data of the current frequency domain feature subset and the current time domain feature subset, and associates the frequency domain features and time domain features belonging to the same current waveform according to the preset feature classification rules to form a complete data set containing all features. This set is the current feature set of the load circuit.

[0034] Furthermore, when evaluating the steady-state characteristics of the voltage waveform, a steady-state monitoring circuit is used. This circuit continuously collects voltage amplitude data when the voltage waveform is in a stable operating phase, calculates the fluctuation range and average deviation of these amplitude data, and determines the stability of the voltage amplitude based on the magnitude of the fluctuation range and average deviation, thus forming the amplitude stability characteristics of the voltage waveform. At the same time, the circuit mirrors and compares the positive and negative half-cycles of the voltage waveform, checking the degree of matching between the two in terms of shape, amplitude, and duration. Based on the degree of matching, the symmetry of the waveform is determined, forming the waveform symmetry characteristics. The amplitude stability characteristics and waveform symmetry characteristics are integrated together to form a subset of the steady-state voltage characteristics of the load circuit.

[0035] Furthermore, when performing dynamic process tracking of the voltage waveform, a dynamic response monitoring circuit is employed. This circuit begins operating the instant the voltage waveform undergoes a sudden change, recording the transition process from the initial state to the instantaneous state after the change, including information such as the rate, amplitude, and duration of the change, thus forming the transient response characteristics of the voltage waveform. After the transient process ends, the circuit continues to monitor the entire process of the voltage waveform recovering to a stable state, recording information such as the time required for recovery, fluctuations during the recovery process, and whether it eventually returns to a stable value, thus forming recovery characteristic characteristics. The transient response characteristics and recovery characteristic characteristics are integrated together to form a subset of the voltage dynamic characteristics of the load circuit.

[0036] Furthermore, when merging the voltage steady-state feature subset and the voltage dynamic feature subset, this is done through a feature aggregation unit. This unit receives all feature data from the voltage steady-state feature subset and the voltage dynamic feature subset, and arranges and associates the features of the steady-state stage and the features of the dynamic stage in an orderly manner according to the complete operating cycle of the voltage waveform, forming a comprehensive data set that covers all operating state features of the voltage waveform. This set is the voltage feature set of the load circuit.

[0037] In general, frequency domain feature extraction of current waveforms involves identifying the main frequency components and their distribution in the current waveform through specific frequency domain analysis methods, statistically analyzing the concentration of spectral energy, and then obtaining the distribution characteristics of the main frequency components and the concentration characteristics of spectral energy of the current waveform. These two features are then integrated as a subset of the current frequency domain features of the load circuit.

[0038] In general, time-varying characteristic analysis of current waveforms involves continuously monitoring the trend of current waveform amplitude change over time, comparing the differences and changes between the actual waveform and the standard waveform, obtaining the amplitude envelope characteristics and waveform distortion evolution characteristics of the current waveform, and integrating these two characteristics as a subset of the current time-domain characteristics of the load circuit.

[0039] In summary, feature integration of the current frequency domain feature subset and the current time domain feature subset involves associating and integrating all current-related features contained in the two subsets to form a complete current feature set, which is the current feature set of the load circuit.

[0040] In general, the steady-state characteristic assessment of voltage waveform involves analyzing the amplitude fluctuations and comparing the matching between the positive and negative half-cycles of the waveform when the voltage waveform is in a stable operating phase. This yields the amplitude stability characteristics and waveform symmetry characteristics of the voltage waveform, and these two characteristics are then integrated as a subset of the steady-state voltage characteristics of the load circuit.

[0041] In summary, dynamic process tracking of voltage waveforms involves recording the transition from the initial state to the instantaneous state after the change, as well as the entire process of recovery to a steady state, when the voltage waveform changes. This yields the transient response characteristics and recovery characteristics of the voltage waveform, and these two characteristics are integrated as a subset of the voltage dynamic characteristics of the load circuit.

[0042] In summary, merging the steady-state voltage feature subset and the dynamic voltage feature subset involves systematically associating and integrating all voltage-related features contained in the two subsets according to the voltage operating cycle to form a complete voltage feature set, which is the voltage feature set of the load circuit.

[0043] S3, the current feature set and the voltage feature set are correlated and fused to obtain the power quality feature sequence of the load circuit; In this embodiment of the invention, the step of correlating and fusing the current feature set and the voltage feature set to obtain the power quality feature sequence of the load circuit includes: The current feature set and voltage feature set are time-aligned to obtain the time calibration feature set of the load circuit. Cross-correlation analysis is performed on the time calibration feature set to obtain the feature correlation matrix of the load circuit; The feature correlation matrix is ​​reconstructed in a structured manner to obtain the current-voltage feature mapping table of the load circuit; Based on the time dimension set and the feature dimension set, the current-voltage feature mapping table is serialized and recombined to obtain the power quality feature sequence of the load circuit.

[0044] The cross-correlation analysis of the time calibration feature set to obtain the feature correlation matrix of the load loop includes: Obtain the cross-correlation coefficient between the current feature set and the voltage feature set in the time domain to obtain the time-domain correlation index of the load circuit; Frequency domain characteristic analysis is performed on the current feature set and the voltage feature set to obtain the frequency domain correlation index of the load circuit; The time-domain correlation index and the frequency-domain correlation index are correlated and fused to obtain the feature correlation matrix of the load loop.

[0045] The step of serializing and reorganizing the current-voltage feature mapping table based on the time dimension set and the feature dimension set to obtain the power quality feature sequence of the load circuit includes: Based on the acquisition time sequence, the feature data in the current-voltage feature mapping table are arranged in time sequence to obtain the time dimension sequence of the load circuit. The feature data are sorted in a secondary order according to the importance of the features to obtain the feature dimension sequence of the load loop; By doubly indexing the time dimension sequence and the feature dimension sequence, the power quality feature sequence of the load circuit is obtained; Specifically, when performing time-series alignment processing on the current feature set and voltage feature set, a timestamp synchronization unit is used. This unit pre-records the acquisition timestamp of each feature in the current feature set and the acquisition timestamp of each feature in the voltage feature set. Then, the current features and voltage features are matched one by one according to the acquisition timestamp to ensure that the current features and voltage features acquired at the same time correspond one-to-one. For isolated features that do not have the same timestamp, they are directly discarded. The successfully matched current-voltage feature combinations are arranged in chronological order to obtain the time calibration feature set of the load circuit.

[0046] Furthermore, when performing cross-correlation analysis on the time calibration feature set, a correlation detection circuit is used. This circuit receives the current features and corresponding voltage features in the time calibration feature set, continuously monitors the changing trends of the current features and voltage features within the same time combination, judges the consistency of the changing direction and the degree of correlation of the changing magnitude of the two, and records the correlation of each set of current-voltage features in a structured form. The set of records of all correlation cases is the feature correlation matrix of the load circuit.

[0047] Furthermore, the structured reconstruction of the feature correlation matrix is ​​achieved through a feature reconstruction unit. This unit first analyzes the physical meaning of each current feature and each voltage feature in the feature correlation matrix. Then, based on the correlation of physical meaning, it classifies the current features and voltage features in the matrix into pairs and presents the correspondence of each pair of correlated features in tabular form to obtain the current-voltage feature mapping relationship table of the load circuit.

[0048] Furthermore, when serializing and recombining the current-voltage feature mapping table based on the time dimension set and the feature dimension set, a sequence recombination unit is used. This unit first reads the time dimension set and arranges the feature pairs in the current-voltage feature mapping table in order of acquisition time from earliest to latest. Then, it reads the feature dimension set and arranges the feature pairs in order of importance from highest to lowest within the same acquisition time. Finally, all the feature pairs arranged in time order and importance order are sequentially connected in series to form a continuous feature sequence, which is the power quality feature sequence of the load circuit.

[0049] Specifically, when obtaining the cross-correlation coefficient between the current feature set and the voltage feature set in the time domain, a time-domain synchronous monitoring circuit is used. This circuit simultaneously receives the current feature and the corresponding voltage feature from the time-calibrated feature set, continuously tracks the amplitude fluctuations of both over time, records the direction of change of the voltage feature amplitude when the current feature amplitude rises, the corresponding state of the voltage feature amplitude when the current feature amplitude reaches its peak, and the degree of overlap of the fluctuation periods of the two. By comparing the synchronicity of these time-domain changes, a parameter reflecting the degree of time-domain correlation between the two is determined. This parameter is the time-domain correlation index of the load circuit.

[0050] Furthermore, when performing frequency domain characteristic analysis on the current and voltage feature sets, a frequency domain spectrum comparison circuit is used. This circuit converts the current features in the time calibration feature set into frequency domain signals, identifies the frequency components with high energy proportions, and simultaneously converts the corresponding voltage features into frequency domain signals, identifies their main frequency components, and then compares the overlap of the main frequency components of the two, as well as the energy change trend of the same frequency component in the current and voltage features. Based on the degree of matching of these frequency domain characteristics, parameters are determined, which are the frequency domain correlation index of the load circuit.

[0051] Furthermore, when fusing the time-domain correlation index and the frequency-domain correlation index, it is achieved through a correlation integration unit. This unit receives the time-domain correlation index and the frequency-domain correlation index corresponding to the same set of current and voltage characteristics, makes a comprehensive judgment on the correlation reflected by the two indicators, and fills the comprehensive correlation judgment results of all current and voltage characteristics into the preset matrix structure one by one according to the order of current and voltage characteristics. The resulting matrix is ​​the characteristic correlation matrix of the load circuit.

[0052] Specifically, when arranging the feature data in the current-voltage feature mapping table according to the acquisition time sequence, a time stamp reading and sorting circuit is used. This circuit is clock-synchronized with the signal acquisition module of the vacuum pump load circuit and can accurately read the time stamps recorded synchronously when each set of feature data in the current-voltage feature mapping table is acquired. Then, the circuit arranges each set of feature data in the order of the time stamps from early to late, ensuring that the feature data acquired earlier is at the beginning of the sequence and the feature data acquired later is at the end of the sequence, and the current feature and voltage feature in each set of feature data always maintain a corresponding relationship. The ordered set of feature data formed after this sorting process is the time dimension sequence of the load circuit.

[0053] Furthermore, when performing secondary sorting of feature data according to the degree of feature importance, a feature importance determination and sorting unit is adopted. This unit pre-stores feature importance rules that meet the requirements of power measurement. Then, the unit reads the specific type of each group of feature data in the current-voltage feature mapping table, determines the importance level of each group of feature data one by one according to the preset feature importance rules, and then arranges all feature data in order from high to low importance level. During the arrangement process, the correspondence between current features and voltage features in each group of feature data remains unchanged. The ordered feature data set formed after this sorting process is the feature dimension sequence of the load circuit.

[0054] Furthermore, when performing dual indexing on the time dimension sequence and the feature dimension sequence, a dual indexing integration circuit is used. This circuit first receives the time dimension sequence and, according to the chronological order of the time dimension sequence, decomposes a set of feature data corresponding to each time node. Then, for the feature data group under each time node, the circuit reads the importance ranking order of the corresponding type of feature data in the feature dimension sequence and rearranges the feature data in the feature data group of the current time node in descending order of importance to ensure that the feature data under each time node meets the feature importance ranking requirements. After that, the circuit concatenates the feature data under all time nodes that have completed the importance ranking in chronological order to form a continuous feature sequence that simultaneously contains the chronological order and the feature importance order, and the correspondence between each group of current features and voltage features is clear. This sequence is the power quality feature sequence of the load circuit.

[0055] In summary, time-alignment processing of the current and voltage feature sets involves matching their acquisition time information to ensure accurate correspondence between current and voltage features at the same moment, eliminating isolated features without corresponding times, and forming a time-calibrated feature set of load circuits arranged in time order. This lays the foundation for time synchronization in subsequent power feature correlation analysis.

[0056] In summary, cross-correlation analysis of the time calibration feature set involves analyzing the correlation between the current and voltage characteristics at the same time in terms of their changing trends and performance. This analysis determines the set of parameters that reflect the correlation between the two, forming a feature correlation matrix for the load circuit, and providing a basis for clarifying the intrinsic relationship between current and voltage characteristics.

[0057] In summary, the structured reconstruction of the feature correlation matrix is ​​based on the physical correlation between current and voltage features. The correlation information in the matrix is ​​organized into pairs of current-voltage feature correspondences, forming a current-voltage feature mapping table for the load circuit, making the correspondence between current and voltage features more intuitive and clear.

[0058] In summary, the serialization and reorganization of the current-voltage feature mapping table based on the time dimension set and feature dimension set involves sorting and integrating the feature data in the mapping table according to the acquisition time sequence and the importance of the features, forming a continuous feature sequence that simultaneously reflects the time sequence and the importance of the features. This sequence is the power quality feature sequence of the load circuit, providing a complete feature basis for accurate power measurement and analysis.

[0059] In summary, obtaining the cross-correlation coefficients of the current feature set and the voltage feature set in the time domain to obtain the time-domain correlation index of the load circuit is achieved by analyzing the amplitude changes, fluctuation synchronicity, and peak occurrence timing of the two in the time dimension. This determines an index that reflects the degree of correlation between them in the time domain, and provides a time-domain basis for subsequent power feature correlation analysis.

[0060] In summary, frequency domain characteristic analysis of current and voltage feature sets to obtain the frequency domain correlation index of the load circuit involves converting both into frequency domain signals, comparing the overlap of major frequency components, the energy change trend of the same frequency component, and other frequency domain characteristics to determine an index that reflects their matching degree at the frequency domain level. This index provides key information in the frequency domain dimension for subsequent energy feature correlation analysis.

[0061] In summary, fusing time-domain correlation indicators and frequency-domain correlation indicators to obtain the characteristic correlation matrix of the load circuit involves comprehensively integrating the time-domain and frequency-domain correlations reflected by the two indicators, and organizing them into a structured matrix according to the correspondence between current characteristics and voltage characteristics. This matrix clearly presents the comprehensive correlation status of current and voltage characteristics, providing a core correlation basis for the subsequent construction of power quality characteristic sequences.

[0062] In summary, arranging the feature data in the current-voltage feature mapping table according to the acquisition time sequence to obtain the time dimension sequence of the load circuit involves sorting each group of current-voltage feature data in the mapping table according to the actual acquisition order, ensuring that the feature data is consistent with the power operation sequence of the vacuum pump load circuit, and laying the foundation for the subsequent construction of a power quality feature sequence with time continuity.

[0063] In summary, the secondary sorting of feature data based on the importance of features to obtain the feature dimension sequence of the load circuit combines the influence of different features in power measurement on the accuracy of energy consumption calculation. The feature data in the mapping table are sorted from high to low importance, and key feature information for power quality assessment is retained first, providing an orderly basis for the subsequent accurate extraction of core power features.

[0064] In summary, the dual indexing of the time dimension sequence and the feature dimension sequence to obtain the power quality feature sequence of the load circuit is to integrate the feature data sorted by time and the feature data sorted by importance. This not only ensures the temporal continuity of the feature data, but also highlights the priority of the core features, and finally forms an ordered feature set that can comprehensively reflect the power operation status of the vacuum pump load circuit.

[0065] S4. Based on the power quality characteristic sequence, perform feature adaptation correction on the current waveform and the voltage waveform to obtain the effective values ​​of the current and voltage of the load circuit. In this embodiment of the invention, the step of performing feature adaptation correction on the current waveform and the voltage waveform based on the power quality feature sequence to obtain the effective current value and effective voltage value of the load circuit includes: The power quality feature sequence is matched and compared with a preset benchmark feature library to obtain the feature deviation pattern of the load circuit. Based on the characteristic deviation mode, the current waveform and the voltage waveform are parametrically adjusted to obtain the current waveform correction parameters and voltage waveform correction parameters of the load circuit; Based on the current waveform correction parameters and the voltage waveform correction parameters, the current waveform and the voltage waveform are dynamically reconstructed to obtain the reconstructed current waveform and the reconstructed voltage waveform of the load circuit. The reconstructed current waveform and the reconstructed voltage waveform are converted to RMS values ​​to obtain the RMS values ​​of the current and voltage of the load circuit.

[0066] The step of matching and comparing the power quality feature sequence with a preset reference feature library to obtain the feature deviation pattern of the load circuit includes: Based on a pre-set benchmark feature library, the power quality feature sequence is evaluated for similarity to obtain a similarity metric for the load circuit. The calculation formula for the similarity metric is as follows: ; In the formula, The similarity metric value, The first in the power quality characteristic sequence One feature parameter, The first standard feature in the benchmark feature library One feature parameter, The average value of the characteristic parameters of the power quality characteristic sequence. The average value of the standard characteristic parameters. Indicates the total number of feature parameters; Based on the similarity measurement results, the deviation feature quantity of the power quality feature sequence relative to the benchmark feature template is extracted; The distribution characteristics of the deviation features are analyzed to identify the characteristic deviation patterns of the load circuit.

[0067] Specifically, when matching and comparing the power quality feature sequence with a pre-set benchmark feature library, a feature comparison unit is used. This unit pre-stores a benchmark feature library that conforms to the normal operating conditions of the vacuum pump load circuit. The benchmark feature library contains complete information directly related to power quality, such as the standard amplitude range, standard waveform shape, and standard change trend of the current and voltage characteristics of the vacuum pump under different typical operating conditions. Moreover, the feature data in the library has been verified by long-term field tests to ensure a high degree of consistency with the actual normal operating conditions of the vacuum pump. Subsequently, the feature comparison unit reads each feature data in the power quality feature sequence, including specific feature items such as current amplitude, voltage waveform curvature, current change rate, and voltage phase offset at different times. At the same time, it retrieves the standard feature data corresponding to the current operating condition from the benchmark feature library. For each feature item, a detailed comparison is performed point by point and time by time from three dimensions: amplitude, waveform details, and change pattern.

[0068] Furthermore, during the comparison process, the location, type, and degree of each difference are recorded in real time. These differences are then categorized and summarized according to time sequence and deviation type, forming a set with a clear structure and complete content that comprehensively reflects the deviation of the power quality characteristic sequence from the standard characteristics in various dimensions. This set is the characteristic deviation pattern of the load circuit.

[0069] Furthermore, when parametrically adjusting the current and voltage waveforms based on the characteristic deviation mode, a waveform parameter adjustment unit is used. This unit first receives the characteristic deviation mode and then analyzes all the deviation information in the deviation mode one by one through the built-in deviation identification module to clarify the target waveform and specific deviation type corresponding to each deviation. Common deviation types include amplitude offset, waveform distortion, and trend deviation.

[0070] Furthermore, for amplitude offset deviations, the waveform parameter adjustment unit combines the deviation amplitude recorded in the deviation mode with the rated current and voltage parameters of the vacuum pump under current operating conditions to calculate the specific amount of compensation needed for the original waveform amplitude. If the amplitude is too low, the unit calculates the amount of upward compensation needed to ensure that the compensated amplitude falls within the standard range under rated operating conditions; if the amplitude is too high, the unit calculates the amount of downward reduction needed to avoid exceeding the rated limit. For waveform distortion deviations, the unit calculates the amplitude and shape adjustment parameters needed to correct the distorted waveform segment based on the start and end times and distortion morphology recorded in the deviation mode, combined with the morphological parameters of the standard waveform in the corresponding time period.

[0071] Furthermore, for trend-related deviations, the unit calculates the required trend slope parameter based on the trend deviation rate recorded in the deviation mode and the rate parameter of the standard trend. These calculated amplitude compensation values, waveform shape adjustment parameters, and trend slope parameters correspond to the current waveform correction parameters and voltage waveform correction parameters that form the load loop, respectively, ensuring that each correction parameter accurately corresponds to a deviation.

[0072] Furthermore, when dynamically reconstructing the current and voltage waveforms based on the current waveform correction parameters and voltage waveform correction parameters, a waveform dynamic reconstruction unit is employed. This unit incorporates a waveform acquisition module, a parameter parsing module, a waveform adjustment module, and a waveform verification module. First, the waveform acquisition module acquires the real-time original current and voltage waveforms of the vacuum pump load circuit, ensuring that the acquired waveform data is consistent with the waveform data source used for feature extraction, thus avoiding the impact of waveform acquisition deviations on the reconstruction effect. Subsequently, the parameter parsing module receives the current waveform correction parameters and voltage waveform correction parameters, breaks down the parameters according to the deviation type and corresponding time period, and clarifies the specific parameter items that need to be adjusted for the original waveform in each time period.

[0073] Furthermore, regarding the original current waveform, the waveform adjustment module adjusts the amplitude of the original current waveform moment by moment according to the correction parameters of the split current waveform. During the time period of amplitude deviation, the current amplitude is increased or decreased in real time according to the amplitude compensation value. During the time period of waveform distortion, the waveform curvature and phase of the distorted segment are changed according to the morphology adjustment parameters to correct the distorted waveform segment into a curve that conforms to the standard morphology. During the time period of trend deviation, the rate of change of the current waveform is adjusted according to the trend slope parameter to make the current change trend conform to the standard trend.

[0074] Furthermore, the adjustment logic for the original voltage waveform is consistent with that for the current waveform. Specifically, the amplitude is adjusted moment-by-moment according to the voltage waveform correction parameters, distortion segments are corrected, and the trend of change is optimized. During the adjustment process, the waveform verification module monitors the continuity and stability of the reconstructed waveform in real time. If a discontinuity or instability is detected at any point, it is immediately fed back to the waveform adjustment module, which then fine-tunes the correction parameters at the corresponding position until the waveform meets the requirements for continuity and stability. After complete adjustment and verification, the reconstructed current and voltage waveforms of the load circuit are finally obtained, accurately reflecting the true electrical energy state of the load circuit.

[0075] Furthermore, when performing RMS conversion on the reconstructed current and voltage waveforms, an RMS conversion unit is used. This unit consists of three parts: a full-wave rectifier circuit, a low-pass filter circuit, and a signal sampling circuit. The parameters of each circuit are designed to be adapted to the rated current and voltage range of the vacuum pump load circuit. For the reconstructed current waveform, it first enters the full-wave rectifier circuit. This circuit adopts a bridge rectifier structure and consists of four rectifier diodes with identical performance. After the reconstructed current waveform enters the circuit, two diodes conduct during the positive half-cycle, and the current flows through the load along a specific path. During the negative half-cycle, the other two diodes conduct, and the current still flows through the load in the same direction. In this way, the alternating positive and negative AC reconstructed current waveform is converted into a unidirectional pulsating DC waveform, ensuring that all amplitude information of the waveform is preserved.

[0076] Furthermore, the unidirectional pulsating DC waveform enters a low-pass filter circuit. This circuit employs an RC filter structure, with the resistors and capacitors selected based on the period of the reconstructed current waveform. The high-frequency pulsating components in the waveform are blocked by the capacitors and consumed by the resistors, ultimately outputting a stable DC signal. The magnitude of this DC signal is equal to the average amplitude of the reconstructed current waveform over one period. From an energy equivalence perspective, this average amplitude is exactly equal to the effective current value of the reconstructed current waveform. Similarly, the reconstructed voltage waveform passes sequentially through a full-wave rectifier circuit and a low-pass filter circuit with the same structure, converting the AC reconstructed voltage waveform into a stable DC signal. The value of this DC signal is the effective voltage value of the reconstructed voltage waveform. Finally, the signal sampling circuit accurately samples the two output DC signals, and the obtained sampled values ​​are used as the effective current and effective voltage values ​​of the load circuit, respectively, providing core electrical parameter data for the subsequent accurate measurement of vacuum pump energy consumption.

[0077] Specifically, when evaluating the similarity of power quality feature sequences based on a pre-set reference feature library, a similarity evaluation circuit is used. This circuit pre-stores a reference feature library that matches the normal operating conditions of the vacuum pump load circuit. The library contains the standard amplitude range, waveform morphology characteristics, and trend characteristics of current and voltage features under various typical operating conditions. After receiving the power quality feature sequence, the circuit retrieves the reference feature template corresponding to the current operating condition from the reference feature library. It compares each feature data in the feature sequence with the standard feature data at the corresponding position in the reference feature template point by point. By detecting the degree of matching between the two in terms of amplitude deviation, waveform morphology overlap, and trend consistency, the similarity of each feature is determined. The similarity of all feature points is integrated to form a parameter set that reflects the overall feature matching. This set is the similarity measure of the load circuit.

[0078] Furthermore, when extracting the deviation features of the power quality feature sequence relative to the benchmark feature template based on the similarity measurement results, a deviation extraction unit is used. This unit receives the similarity measurement results, first identifies the regions with low similarity values, and then extracts the specific differences between the feature data in the feature sequence and the standard data in the benchmark feature template for each feature item corresponding to these regions. For amplitude-related features, the difference between the amplitude of the feature sequence and the standard amplitude is calculated; for waveform morphology-related features, the start and end positions of waveform distortion and the distortion morphology are recorded; for trend-related features, the degree of deviation between the trend slope and the standard slope is determined. These extracted specific difference information are organized according to feature type and time order to form a feature set containing all the details of the differences. This set is the deviation feature of the load circuit.

[0079] Furthermore, when performing pattern recognition on the distribution characteristics of deviation features, a pattern recognition circuit is employed. This circuit receives the deviation features and first analyzes their distribution over time to determine the time periods in which deviations are concentrated and the density of deviations within each time period. Simultaneously, it analyzes the distribution of deviation features by feature type, statistically analyzing the quantity and proportion of amplitude-type, waveform-type, and trend-type deviations. Based on the combined results of the time and type distributions, the deviation features are categorized into different deviation patterns. The characteristics of each pattern are summarized and described, and the resulting structured set of deviation types represents the characteristic deviation patterns of the load circuit.

[0080] Specifically, The similarity metric representing the load circuit is a calculated result that reflects the degree of similarity between the power quality feature sequence and the baseline feature template. Represents a single feature data in a power quality feature sequence, each This corresponds to a specific characteristic of power quality, such as the current amplitude characteristics or voltage waveform curvature characteristics at a certain moment. Represents all of the power quality characteristic sequences The average value is obtained by summing all the feature data in the power quality feature sequence and then dividing by the total number of feature data. Each represents a single standard feature data in the baseline feature template. and The corresponding feature items are standard feature data of vacuum pumps under normal operating conditions stored in the benchmark feature library. Represents all in the baseline feature template The average value is obtained by adding up all the standard feature data in the baseline feature template and then dividing by the total number of standard feature data. This represents the total number of feature data points involved in the calculation, and this number is related to the number of feature data points in the power quality feature sequence. Total number of reference feature templates The total number is exactly the same, ensuring and It can perform calculations in a one-to-one correspondence.

[0081] In summary, matching and comparing the power quality feature sequence with a pre-set reference feature library to obtain the characteristic deviation pattern of the load circuit involves identifying the type and degree of deviation in amplitude, waveform shape, and trend by comparing the differences between the power quality feature sequence and the standard features in the reference feature library. This forms a structured set of deviation patterns, providing a clear basis for subsequent waveform correction.

[0082] In summary, parameterized waveform adjustment of current and voltage waveforms based on characteristic deviation modes to obtain current and voltage waveform correction parameters for the load circuit is achieved by calculating specific parameters for adjusting the current and voltage waveforms based on the deviation type and degree recorded in the characteristic deviation mode. This ensures that each parameter accurately corresponds to a deviation, providing an executable adjustment basis for waveform reconstruction.

[0083] In summary, the reconstructed current and voltage waveforms of the load circuit are dynamically reconstructed based on the current and voltage waveform correction parameters. This involves adjusting the amplitude, shape, and trend of the original current and voltage waveforms at each moment according to the correction parameters, while verifying the continuity and stability of the waveforms. Ultimately, a reconstructed waveform that truly reflects the electrical energy state of the load circuit is obtained, providing a reliable waveform basis for RMS conversion.

[0084] In summary, the effective value conversion of reconstructed current and voltage waveforms to obtain the effective values ​​of current and voltage in the load circuit involves processing the AC reconstructed waveforms through circuits such as rectification and filtering. This process converts the AC reconstructed waveforms into DC signals that reflect their average amplitude. The values ​​of these DC signals are the corresponding effective values, providing core electrical parameters for applications such as load circuit energy consumption metering. In summary, the similarity assessment of power quality feature sequences based on a pre-set benchmark feature library to obtain a similarity metric for load circuits involves comparing each feature data in the power quality feature sequence with the standard feature data of the corresponding operating conditions in the benchmark feature library point by point. The analysis focuses on the degree of agreement between the two in terms of amplitude, waveform shape, and trend of change, and comprehensively forms a metric result that reflects the overall similarity. This provides a quantitative basis for judging the degree of agreement between power quality features and standard features.

[0085] In summary, extracting deviation features of power quality feature sequences relative to the benchmark feature template based on similarity measurement results involves accurately locating feature items in the power quality feature sequence that do not match the benchmark feature template based on the difference regions reflected in the similarity measurement, extracting specific difference information of these feature items in terms of amplitude, waveform, trend, etc., forming a feature set containing all deviation details, and providing specific objects for subsequent deviation pattern identification.

[0086] In summary, pattern identification of the distribution characteristics of deviation features to obtain the characteristic deviation patterns of the load circuit involves analyzing the concentrated time periods of deviation features and the distribution ratio of deviation features in terms of feature types. Based on these distribution characteristics, deviations are classified into different pattern types, clarifying the characteristics and occurrence rules of each pattern, forming a structured set of deviation patterns, and providing a clear direction for subsequent characteristic adaptation correction of current and voltage waveforms.

[0087] S5, integrate the effective value of the current and the effective value of the voltage to obtain the accurate electrical energy measurement value of the vacuum pump.

[0088] In this embodiment of the invention, the step of integrating the effective value of the current and the effective value of the voltage to obtain the accurate electrical energy measurement value of the load circuit includes: The effective values ​​of current and voltage are paired and combined to obtain the electrical parameter data set of the load circuit; The integrity of the electrical parameter data set is verified to obtain the set of electrical parameters that have passed the verification of the load circuit. Based on the preset data structure and storage format, the verified electrical parameter set is standardized and encapsulated to obtain the standardized electrical parameter dataset of the load circuit. By comparing and analyzing the standardized electrical parameter dataset with historical electrical parameter benchmarks, the characteristic patterns and trends of the standardized electrical parameter data are obtained. By integrating the aforementioned characteristic patterns and trends, a power quality report for the load circuit is obtained. The power quality report is used as the accurate power measurement value of the vacuum pump.

[0089] Specifically, when performing electrical parameter pairing and combination of current RMS and voltage RMS values, an electrical parameter pairing unit is used. This unit pre-reads the acquisition time markers corresponding to the current RMS and voltage RMS values, and then matches the current RMS and voltage RMS values ​​acquired at the same time according to the acquisition time markers. This ensures that each pairing data can reflect the current and voltage status of the load circuit at the same time. At the same time, isolated RMS data without corresponding time markers are discarded. The successfully paired current RMS and voltage RMS values ​​are organized in the order of acquisition time to form a set containing multiple sets of synchronous current and voltage data. This set is the electrical parameter data set of the load circuit.

[0090] Furthermore, when performing integrity verification on the electrical parameter data set, an electrical parameter verification unit is used. This unit first checks whether each data set in the electrical parameter data set contains both effective current and effective voltage values. If a data set contains only a single effective value, it is determined to be incomplete and discarded. Subsequently, the unit retrieves the rated electrical parameter range of the vacuum pump load circuit and compares all effective current and effective voltage values ​​in the electrical parameter data set with the rated range. If an effective value exceeds the rated range, it further checks whether there are any special operating conditions in the load circuit at that moment. If there are no special operating conditions, the data is determined to be abnormal and discarded. After integrity and rationality checks, the set of remaining compliant electrical parameter data is the set of electrical parameters that have passed the verification of the load circuit.

[0091] Furthermore, when standardizing and encapsulating the verified electrical parameter set based on the preset data structure and storage format, a standardized encapsulation unit is used. This unit pre-stores the preset data structure and storage format that conform to the data specifications in the field of power measurement. Subsequently, the unit fills each set of data in the verified electrical parameter set into the preset data structure in the field order of "acquisition time - current RMS value - voltage RMS value". At the same time, the data is encoded according to the storage format requirements, and field separators and identifiers are added to ensure that the encapsulated data is completely uniform in structure and format. The resulting standardized data set is the standardized electrical parameter dataset of the load circuit.

[0092] Furthermore, when comparing and analyzing the standardized electrical parameter dataset with historical electrical parameter benchmarks, an electrical parameter comparison and analysis unit is used. This unit pre-stores historical electrical parameter benchmarks under normal operating conditions of the load circuit. Subsequently, the unit compares each set of data in the standardized electrical parameter dataset with the data of the same period in the historical electrical parameter benchmark, analyzes the degree of consistency between the current data and the historical benchmark, and then determines the characteristic pattern corresponding to the current data. At the same time, the unit tracks the changes in the effective values ​​of current and voltage in the order of acquisition time, observes whether the data shows a gradual increase, a gradual decrease, or a stable state, forms the trend of data change, and finally obtains the characteristic pattern and trend of standardized electrical parameter data.

[0093] Furthermore, when integrating characteristic patterns and trends, a report integration unit is used. This unit first classifies and describes the characteristic patterns, clarifies the current power operation mode of the load circuit, and explains the basis for the mode determination. Then, it elaborates on the trends in detail, including the trend type and the time period in which the trend occurs. At the same time, the unit will include key data from the verified electrical parameter set into the integration content and organize them into a document according to the logical structure of "data overview - characteristic pattern analysis - trend analysis". This document is the power quality report of the load circuit.

[0094] Furthermore, when using the power quality report as the precise power measurement value of the vacuum pump, based on the characteristics of the power quality report, which includes RMS current and RMS voltage data that have undergone integrity verification, as well as characteristic patterns and trends reflecting the power operation status, it can comprehensively and accurately reflect the power consumption and operating status of the vacuum pump load circuit. Therefore, the power quality report is directly determined as the precise power measurement value of the vacuum pump, providing a core basis for vacuum pump energy consumption management and operating status monitoring.

[0095] In summary, pairing and combining the RMS current and RMS voltage values ​​to obtain the electrical parameter data set of the load circuit involves matching the acquisition time stamps of both values ​​to ensure a precise correspondence between the RMS current and RMS voltage values ​​at the same moment. This process eliminates isolated invalid data and forms a time-ordered set of electrical parameter combinations, providing synchronously correlated foundational data for subsequent power data processing, in accordance with G01R22. The field requires the integration of correlations between electrical energy parameters.

[0096] In summary, performing integrity checks on electrical parameter data sets to obtain a valid set of electrical parameters for the load circuit involves verifying that each data set includes both effective current and effective voltage values, checking that the data falls within the reasonable range of the load circuit's rated electrical parameters, and eliminating incomplete or abnormal data. This ensures the reliability and reasonableness of the retained data, aligning with G01R22. The core requirement of the field is the accuracy of electrical energy measurement data.

[0097] In summary, the standardized electrical parameter dataset of the load circuit is obtained by standardizing and encapsulating the verified electrical parameter set based on the preset data structure and storage format. The data is organized according to a unified field order and encoding format, so that the data structure is standardized and the format is uniform, which facilitates subsequent data comparison and reading.

[0098] In summary, comparing and analyzing standardized electrical parameter datasets with historical electrical parameter benchmarks to obtain the characteristic patterns and trends of standardized electrical parameter data involves determining the current data's operating mode by comparing the degree of agreement between the current data and historical normal operating condition data, and tracking the data's changes over time, thus providing a basis for power operation status analysis.

[0099] In summary, the power quality report for the load circuit is obtained by integrating characteristic patterns and trends. It is a document that comprehensively reflects the power operation status by organizing the data overview, the basis for determining characteristic patterns and the details of trends according to the logical structure, and providing a complete carrier for accurate power measurement values.

[0100] In summary, the power quality report is used as an accurate measurement of the power consumption of vacuum pumps because it contains verified core electrical parameters, clear operating characteristic modes and trends, and can comprehensively and accurately reflect the power consumption and operating status of the vacuum pump load circuit, providing a reliable basis for vacuum pump energy consumption management.

[0101] like Figure 2 The diagram shown is a functional block diagram of a vacuum pump energy consumption real-time monitoring and analysis system provided in an embodiment of the present invention.

[0102] The vacuum pump energy consumption real-time monitoring and analysis system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the vacuum pump energy consumption real-time monitoring and analysis system 100 may include a point signal conditioning and waveform reconstruction module 101, a multi-domain feature analysis module 102, a feature correlation and sequence fusion module 103, a feature adaptive correction module 104, and an energy metering and output module 105. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0103] In this embodiment, the functions of each module / unit are as follows: The point signal conditioning and waveform reconstruction module is used to perform multi-scale filtering on the current signal and voltage signal of the load circuit in the vacuum pump to obtain the current waveform and voltage waveform of the load circuit. The multi-domain feature analysis module is used to perform time-frequency domain feature analysis on the current waveform and time-domain behavior analysis on the voltage waveform to obtain the current feature set and voltage feature set of the load circuit. The feature association and sequence fusion module is used to associate and fuse the current feature set and the voltage feature set to obtain the power quality feature sequence of the load circuit. The feature adaptive correction module is used to perform feature adaptation correction on the current waveform and the voltage waveform based on the power quality feature sequence to obtain the effective values ​​of the current and voltage of the load circuit. The power measurement and output module is used to integrate the effective values ​​of the current and voltage to obtain accurate power measurement values ​​for the load circuit. In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0104] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0105] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0106] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0107] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for real-time monitoring and analysis of vacuum pump energy consumption, characterized in that, The method includes: S1. Perform multi-scale filtering on the current and voltage signals of the load circuit in the vacuum pump to obtain the current waveform and voltage waveform of the load circuit; S2. Perform time-frequency domain feature analysis on the current waveform and time-domain behavior analysis on the voltage waveform to obtain the current feature set and voltage feature set of the load circuit; S3. The current feature set and the voltage feature set are correlated and fused to obtain the power quality feature sequence of the load circuit; S4. Based on the power quality characteristic sequence, perform feature adaptation correction on the current waveform and the voltage waveform to obtain the effective values ​​of the current and voltage of the load circuit; S5. The effective values ​​of the current and voltage are integrated to obtain the accurate electrical energy measurement value of the vacuum pump.

2. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 1, characterized in that, The process of performing multi-scale filtering on the current and voltage signals of the load circuit in the vacuum pump to obtain the current and voltage waveforms of the load circuit includes: The current signal is subjected to power frequency noise separation to obtain the first-scale filtered current signal of the load circuit; The first-scale filtered current signal is subjected to high-frequency oscillation suppression to obtain the second-scale filtered current signal of the load circuit. Baseline drift correction is performed on the second-scale filtered current signal to obtain the current waveform of the load circuit; The voltage signal is subjected to transient pulse filtering to obtain the preprocessed voltage signal of the load circuit; Common-mode noise suppression is performed on the preprocessed voltage signal to obtain the voltage waveform of the load circuit.

3. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 1, characterized in that, The process of performing time-frequency domain feature analysis on the current waveform and time-domain behavior analysis on the voltage waveform to obtain the current feature set and voltage feature set of the load circuit includes: Frequency domain features are extracted from the current waveform to obtain the distribution features of the main frequency components and the spectral energy concentration features of the current waveform, and the distribution features of the main frequency components and the spectral energy concentration features are used as the current frequency domain feature subset of the load circuit; Time-varying feature analysis is performed on the current waveform to obtain the amplitude envelope feature and waveform distortion evolution feature of the current waveform, and the amplitude envelope feature and waveform distortion evolution feature are used as a subset of the current time domain features of the load circuit. The current frequency domain feature subset and the current time domain feature subset are integrated to obtain the current feature set of the load circuit; The voltage waveform is evaluated for steady-state characteristics to obtain the amplitude stability characteristics and waveform symmetry characteristics of the voltage waveform, and the amplitude stability characteristics and waveform symmetry characteristics are used as a subset of the steady-state voltage characteristics of the load circuit. The voltage waveform is dynamically tracked to obtain transient response characteristics and recovery characteristics, and these transient response characteristics and recovery characteristics are used as a subset of the voltage dynamic characteristics of the load circuit. The steady-state voltage feature subset and the dynamic voltage feature subset are merged to obtain the voltage feature set of the load circuit.

4. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 1, characterized in that, The step of correlating and fusing the current feature set and the voltage feature set to obtain the power quality feature sequence of the load circuit includes: The current feature set and voltage feature set are time-aligned to obtain the time calibration feature set of the load circuit. Cross-correlation analysis is performed on the time calibration feature set to obtain the feature correlation matrix of the load circuit; The feature correlation matrix is ​​reconstructed in a structured manner to obtain the current-voltage feature mapping table of the load circuit; Based on the time dimension set and the feature dimension set, the current-voltage feature mapping table is serialized and recombined to obtain the power quality feature sequence of the load circuit.

5. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 4, characterized in that, The cross-correlation analysis of the time calibration feature set to obtain the feature correlation matrix of the load loop includes: Obtain the cross-correlation coefficient between the current feature set and the voltage feature set in the time domain to obtain the time-domain correlation index of the load circuit; Frequency domain characteristic analysis is performed on the current feature set and the voltage feature set to obtain the frequency domain correlation index of the load circuit; The time-domain correlation index and the frequency-domain correlation index are correlated and fused to obtain the feature correlation matrix of the load loop.

6. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 4, characterized in that, The step of serializing and reorganizing the current-voltage feature mapping table based on the time dimension set and the feature dimension set to obtain the power quality feature sequence of the load circuit includes: Based on the acquisition time sequence, the feature data in the current-voltage feature mapping table are arranged in time sequence to obtain the time dimension sequence of the load circuit. The feature data are sorted in a secondary order according to the importance of the features to obtain the feature dimension sequence of the load loop; By performing dual indexing on the time dimension sequence and the feature dimension sequence, the power quality feature sequence of the load circuit is obtained.

7. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 1, characterized in that, The step of performing feature adaptation correction on the current waveform and the voltage waveform based on the power quality characteristic sequence to obtain the effective values ​​of the current and voltage of the load circuit includes: The power quality feature sequence is matched and compared with a preset benchmark feature library to obtain the feature deviation pattern of the load circuit. Based on the characteristic deviation mode, the current waveform and the voltage waveform are parametrically adjusted to obtain the current waveform correction parameters and voltage waveform correction parameters of the load circuit; Based on the current waveform correction parameters and the voltage waveform correction parameters, the current waveform and the voltage waveform are dynamically reconstructed to obtain the reconstructed current waveform and the reconstructed voltage waveform of the load circuit. The reconstructed current waveform and the reconstructed voltage waveform are converted to RMS values ​​to obtain the RMS values ​​of the current and voltage of the load circuit.

8. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 7, characterized in that, The step of matching and comparing the power quality feature sequence with a preset reference feature library to obtain the feature deviation pattern of the load circuit includes: Based on a pre-set benchmark feature library, the power quality feature sequence is evaluated for similarity to obtain a similarity metric for the load circuit. The calculation formula for the similarity metric is as follows: ; In the formula, The similarity metric value, The first in the power quality characteristic sequence One feature parameter, The first standard feature in the benchmark feature library One feature parameter, The average value of the characteristic parameters of the power quality characteristic sequence. The average value of the standard characteristic parameters. Indicates the total number of feature parameters; Based on the similarity measurement results, the deviation feature quantity of the power quality feature sequence relative to the benchmark feature template is extracted; The distribution characteristics of the deviation features are analyzed to identify the characteristic deviation patterns of the load circuit.

9. The method for real-time monitoring and analysis of vacuum pump energy consumption as described in claim 1, characterized in that, The process of integrating the effective values ​​of the current and voltage to obtain the accurate electrical energy measurement value of the load circuit includes: The effective values ​​of current and voltage are paired and combined to obtain the electrical parameter data set of the load circuit; The integrity of the electrical parameter data set is verified to obtain the set of electrical parameters that have passed the verification of the load circuit. Based on the preset data structure and storage format, the verified electrical parameter set is standardized and encapsulated to obtain the standardized electrical parameter dataset of the load circuit. By comparing and analyzing the standardized electrical parameter dataset with historical electrical parameter benchmarks, the characteristic patterns and trends of the standardized electrical parameter data are obtained. By integrating the aforementioned characteristic patterns and trends, a power quality report for the load circuit is obtained. The power quality report is used as the accurate power measurement value of the vacuum pump.

10. A real-time monitoring and analysis system for vacuum pump energy consumption, characterized in that, The system includes: The point signal conditioning and waveform reconstruction module is used to perform multi-scale filtering on the current and voltage signals of the load circuit in the vacuum pump to obtain the current and voltage waveforms of the load circuit. The multi-domain feature analysis module is used to perform time-frequency domain feature analysis on the current waveform and time-domain behavior analysis on the voltage waveform to obtain the current feature set and voltage feature set of the load circuit. The feature association and sequence fusion module is used to associate and fuse the current feature set and the voltage feature set to obtain the power quality feature sequence of the load circuit; The feature adaptive correction module is used to perform feature adaptation correction on the current waveform and the voltage waveform based on the power quality feature sequence to obtain the effective values ​​of the current and voltage of the load circuit. The power metering and output module is used to integrate the effective values ​​of the current and the effective values ​​of the voltage to obtain the accurate power measurement value of the load circuit.

Images