Enterprise operation state monitoring method based on enterprise power data

Abstract

The invention belongs to the technical field of data analysis, and discloses an enterprise operation state monitoring method based on enterprise power data, and the method comprises the steps: carrying out the feature decomposition and modal component analysis of total power time sequence data, and combining a reference power feature set of main production equipment, accurate identification and analysis of the working state of each main production device of an enterprise are realized. According to the method, the limitation that a traditional method only depends on the total electricity consumption is overcome, the actual production condition of an enterprise can be reflected more accurately, manual operation is not prone to occurring, and therefore a more reliable technical means is provided for monitoring the operation state of the enterprise. Real-time monitoring and early warning of the enterprise operation state are realized, the informatization level can be improved, the workload of loan management personnel is reduced, the data is ensured to be real and reliable, and the real-time performance and accuracy of data analysis are improved.

Description

Technical Field

[0001] The present application relates to the technical field of data analysis, and in particular to a method for monitoring enterprise operating status based on enterprise power data. Background Art

[0002] Traditional post-loan management methods of financial institutions usually collect information such as the company's financial data, bank statements, electricity bills, tax bills, corporate legal proceedings, etc., and post-loan management personnel conduct on-site inspections of the company. However, this method has many problems: the degree of digitization of information collection is low, and data collation takes a long time; data information is easy to be falsified; the collected data can often only reflect the company's operating conditions over a long period of time, and cannot reflect the company's current operating status in a timely and accurate manner, let alone early warning of corporate operating risks, causing financial institutions to miss the best time to deal with risks.

[0003] At present, financial institutions have limited means of post-loan management, low efficiency, poor results, and low levels of informatization and digitization. Therefore, it is urgent to find new methods to improve the level of informatization, reduce the workload of loan managers, ensure that data is authentic and reliable, and improve the real-time and accuracy of data analysis.

[0004] The business status of an enterprise is closely related to the order saturation, which in turn directly affects the production and manufacturing activities of the enterprise. The more orders there are, the busier the production is and the more frequently the production equipment is started. Since the operation of production equipment consumes electricity, by monitoring the production electricity consumption data of the enterprise, we can understand the order status of the enterprise and then understand whether the business status of the enterprise is good.

[0005] The commonly used method for judging the business status of an enterprise based on electricity consumption data usually uses the changes and trends of the enterprise's electricity consumption over a period of time to judge the business status of the enterprise. However, this method has the following problems: The amount of electricity consumption data is small and the data dimension is too single. Enterprises may increase electricity consumption by turning on electricity-consuming equipment that is not related to production and operation, disguising it as having sufficient orders and full production load, in order to deceive financial institutions.

[0006] There are many electrical equipments in an enterprise, but not all of them are highly correlated with the enterprise's operating status. The existing enterprise operating status judgment method uses the power consumption information of all electrical equipments for analysis, which leads to unsatisfactory accuracy of the judgment results.

[0007] In addition, existing methods cannot monitor the business status of enterprises in real time, making it difficult to detect potential business risks in a timely manner. At the same time, they lack the ability to predict the future business status of enterprises and cannot provide forward-looking risk management advice to financial institutions.

[0008] In view of the above problems, the existing technology needs to be improved urgently. Summary of the invention

[0009] The purpose of this application is to provide a method for monitoring the business status of an enterprise based on enterprise power data, which can improve the level of informatization, reduce the workload of loan management personnel, ensure the authenticity and reliability of data, and improve the real-time and accuracy of data analysis.

[0010] The present application provides a method for monitoring enterprise operation status based on enterprise power data, comprising the steps of: A1. Obtain the total power consumption time series data of the enterprise, and perform feature decomposition on the total power consumption time series data according to a preset time period length to obtain multiple modal components corresponding to each time period; A2. Obtain a reference power feature set of each major production equipment of the enterprise; the reference power feature set includes a reference time domain feature and a reference frequency domain feature; A3. Determine the validity of each of the modal components according to the characteristic entropy of the modal components, and perform noise filtering on the valid modal components to obtain filtered modal components; A4. extracting the time domain features and frequency domain features of each of the filtered modal components to form an actual power feature set corresponding to each of the filtered modal components, which is used to match the reference power feature set to identify the corresponding relationship between each of the filtered modal components and each of the main production equipment; A5. Obtain the working status data of each of the main production equipment according to the matching identification results corresponding to each of the time periods; A6. Determine the current operating status of the enterprise based on the working status data.

[0011] This method achieves accurate identification and analysis of the working status of each major production equipment of the enterprise by performing feature decomposition and modal component analysis on the total power time series data, combined with the reference power feature set of the main production equipment. This method overcomes the limitation of the traditional method that only relies on total power consumption, can more accurately reflect the actual production situation of the enterprise, and is not easily manipulated by humans, thus providing a more reliable technical means for monitoring the business status of the enterprise. It realizes real-time monitoring and early warning of the business status of the enterprise, which can improve the level of informatization, reduce the workload of loan management personnel, ensure the authenticity and reliability of data, and improve the real-time and accuracy of data analysis.

[0012] Preferably, step A1 comprises: A101. Obtain the total power consumption time series data of the enterprise for a preset time period; A102. Divide the total power time series data according to the time period length to obtain a plurality of total power time series data segments; A103. Perform feature decomposition on each of the total electric power time series data segments based on a modal decomposition algorithm to obtain multiple modal components corresponding to each of the time periods.

[0013] Preferably, in step A103, a variational mode decomposition algorithm is used to perform characteristic decomposition on each of the total electric power time series data segments.

[0014] The variational mode decomposition algorithm is an advanced signal processing technology that can effectively decompose complex time series data into multiple modal components. In solving the problem of characteristic decomposition of enterprise total power time series data, the variational mode decomposition algorithm has the following advantages: it can adaptively determine the number of decomposed modes without manual setting; it has strong robustness to noise and can better extract useful signal features; compared with other modal decomposition methods, such as empirical mode decomposition (EMD), the variational mode decomposition algorithm can better handle mode aliasing problems.

[0015] Preferably, the reference time domain feature includes at least one of a reference mean, a reference standard deviation, a reference kurtosis, a reference mean square value, a reference root mean square value, a reference peak factor, a reference pulse factor, and a reference waveform factor; The reference frequency domain feature includes at least one of a reference center frequency, a reference bandwidth, a reference power spectrum density, a reference frequency variance, and a reference center of gravity frequency; The time domain characteristics include at least one of mean, standard deviation, kurtosis, mean square value, root mean square value, peak factor, pulse factor, and waveform factor; The frequency domain characteristics include at least one of center frequency, bandwidth, power spectrum density, frequency variance, and center of gravity frequency.

[0016] The combined use of these features enables a comprehensive description of the characteristics of power data. Through standardized feature definitions, a reliable feature basis is provided for enterprise operating status monitoring. The time domain features reflect the statistical distribution characteristics of power data, and the frequency domain features reflect the frequency composition characteristics of power data. The combined use of the two types of features can comprehensively describe and compare the characteristics of power data. This enables a more accurate matching of the filtered modal components with the main production equipment.

[0017] Preferably, step A3 comprises: A301. Calculate the characteristic entropy of each of the modal components; A302. Compare the characteristic entropy of each of the modal components with a preset characteristic entropy threshold, and determine the modal component whose characteristic entropy is lower than the characteristic entropy threshold as a valid modal component; A303. Perform noise filtering on the effective modal components to obtain filtered modal components.

[0018] Preferably, in step A303, a Wiener adaptive filter algorithm is used to perform noise filtering processing on the effective modal components.

[0019] Preferably, step A4 comprises: A401. Extracting the time domain features of each of the filtered modal components; A402. Extracting the frequency domain characteristics of each of the filtered modal components based on FFT transform; A403. Using the extracted time domain features and the frequency domain features to form an actual power feature set of each of the filtered modal components; A404. Compare the actual power characteristic set of each of the filtered modal components with the reference power characteristic set of each of the main production equipment, and determine the corresponding relationship between each of the filtered modal components and each of the main production equipment.

[0020] Preferably, the working status data includes start time, working time, stop time and steady-state power value change index; Step A5 includes: According to the filtered modal components corresponding to each of the main production equipment in each of the time periods, identifying the start time, working time and stop time of each of the main production equipment; According to the characteristic entropy of the filtered modal component corresponding to each of the main production equipment in each of the time periods, the steady-state power value variation index of each of the main production equipment is calculated.

[0021] Preferably, step A6 comprises: Calculate an abnormal value of the operating state according to the start time, the working time, the stop time and the steady-state power value change index; Whether the current operating status of the enterprise is abnormal is determined according to the operating status abnormal value.

[0022] Preferably, after step A6, the method further comprises the following steps: A7. If the current operating status of the enterprise is normal, predict the estimated power consumption time series data of the enterprise based on the total power consumption time series data; the estimated power consumption time series data is the estimated data of the total power consumption time series data of the enterprise in the future; A8. Evaluate the potential risk of abnormal operating status of the enterprise based on the estimated power consumption time series data.

[0023] Beneficial effects: The enterprise operating status monitoring method based on enterprise power data provided by this application realizes accurate identification and analysis of the working status of each major production equipment of the enterprise by performing feature decomposition and modal component analysis on the total power consumption time series data, combined with the reference power feature set of the main production equipment. This method overcomes the limitation of the traditional method that only relies on total power consumption, can more accurately reflect the actual production situation of the enterprise, and is not easily manipulated by humans, thus providing a more reliable technical means for monitoring the business status of the enterprise. It realizes real-time monitoring and early warning of the business status of the enterprise, which can improve the level of informatization, reduce the workload of loan management personnel, ensure the authenticity and reliability of data, and improve the real-time and accuracy of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a method for monitoring enterprise operating status based on enterprise power data provided in an embodiment of the present application.

[0025] Figure 2 A flowchart of another method for monitoring enterprise operating status based on enterprise power data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0027] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0028] In the field of enterprise operation status monitoring, traditional methods mainly rely on collecting information such as the enterprise's financial data, bank statements, electricity bills, and on-site inspections by post-loan management personnel. However, this method has problems such as low information collection efficiency, possible data falsification, and information lag. It cannot accurately reflect the current operating status of the enterprise in a timely manner, and it is even more difficult to warn of potential risks in advance.

[0029] In recent years, the method of judging the operating status based on the electricity consumption data of enterprises has been applied. This method evaluates the operating status of enterprises by analyzing the changes and trends in the electricity consumption of enterprises over a period of time. However, this method still faces technical challenges such as single data dimension, easy disguise by enterprises through non-production electricity equipment, and difficulty in distinguishing the electricity consumption of key production equipment. Specifically, the amount of electricity consumption information data alone is small and cannot fully reflect the actual production status of the enterprise. At the same time, enterprises may increase the total electricity consumption by turning on electricity equipment that is not related to production and operation, thereby disguising the illusion of sufficient orders and full production load. In addition, since enterprises usually have a variety of electricity-consuming equipment, but not all equipment is highly relevant to the operating status, the analysis method based on total electricity consumption is difficult to accurately reflect the true operating status of the enterprise.

[0030] If this technical problem cannot be effectively solved, the following serious consequences will occur: First, financial institutions may not be able to detect the deterioration of corporate operating conditions in a timely manner, miss the best time for risk intervention, and increase the risk of loan defaults. Second, due to insufficient data reliability, financial institutions may need to invest more manpower in on-site inspections, increasing management costs. Furthermore, companies may take advantage of the limitations of this monitoring method and manipulate electricity consumption data to conceal their true operating conditions, which not only increases financial risks but may also lead to misallocation of resources.

[0031] In addition, this problem may also affect the efficiency of credit decision-making of financial institutions. Due to the inability to obtain accurate and real-time information on the business status of enterprises, financial institutions may be too conservative or aggressive when approving new loan applications or adjusting existing loan conditions, affecting the effective allocation of financial resources. From a technical perspective, this problem highlights the shortcomings of existing data collection and analysis methods in terms of accuracy, real-time and anti-interference capabilities, and a more intelligent and comprehensive technical solution is urgently needed.

[0032] Based on the analysis of the existing technical problems, this application proposes a new solution.

[0033] Please refer to Figure 1 , Figure 1 A method for monitoring the business status of an enterprise based on enterprise power data in some embodiments of the present application comprises the steps of: A1. Obtain the total power consumption time series data of the enterprise, and perform feature decomposition on the total power consumption time series data according to a preset time period length to obtain multiple modal components corresponding to each time period; A2. Obtain the reference power feature set of each major production equipment of the enterprise; the reference power feature set includes reference time domain features and reference frequency domain features; A3. Determine the validity of each modal component according to the characteristic entropy of the modal component, and perform noise filtering on the valid modal component to obtain the filtered modal component; A4. Extract the time domain features and frequency domain features of each filtered modal component to form an actual power feature set corresponding to each filtered modal component, which is used to match with the reference power feature set to identify the corresponding relationship between each filtered modal component and each major production equipment; A5.According to the matching identification results corresponding to each time period, obtain the working status data of each major production equipment; A6. Determine the company's current operating status based on work status data.

[0034] This method achieves accurate identification and analysis of the working status of each major production equipment of the enterprise by performing feature decomposition and modal component analysis on the total power time series data, combined with the reference power feature set of the main production equipment. This method overcomes the limitation of the traditional method that only relies on total power consumption, can more accurately reflect the actual production situation of the enterprise, and is not easily manipulated by humans, thus providing a more reliable technical means for monitoring the business status of the enterprise. It realizes real-time monitoring and early warning of the business status of the enterprise, which can improve the level of informatization, reduce the workload of loan management personnel, ensure the authenticity and reliability of data, and improve the real-time and accuracy of data analysis.

[0035] Specifically, step A1 includes: A101. Obtain the total power consumption time series data of the enterprise for a preset time period; A102. Divide the total power consumption time series data according to the length of the time period to obtain multiple total power consumption time series data segments; A103. Based on the modal decomposition algorithm, each total power consumption time series data segment is subjected to characteristic decomposition to obtain multiple modal components corresponding to each time period.

[0036] By preprocessing and feature decomposing the total power consumption time series data of the enterprise, the problem of how to extract useful information from complex power consumption data is solved. First, the total power consumption time series data of the preset duration is obtained to ensure the time range of the analysis. Then, the data is divided into segments of multiple time periods so that the analysis can be performed on a finer-grained time scale. Finally, the modal decomposition algorithm is used to feature decompose each data segment to obtain multiple modal components, which may correspond to different production equipment or production activities of the enterprise.

[0037] Among them, the total electricity power time series data refers to the data series of the total electricity power of the enterprise changing with time, which can be specifically collected in real time using smart meters or power monitoring equipment. Here, the total electricity power time series data of a preset time length is collected as the analysis object, and the preset time length can be set according to actual needs, such as one day, one week or one month. It should be noted that the sampling time length of the total electricity power time series data (i.e., the preset time length) does not represent the analysis period (i.e., the time interval for executing the enterprise business status monitoring method based on enterprise power data of this application), and the analysis period can be set according to actual needs. For example, when the preset time length is one month, the analysis period can be one day or one week, but is not limited to this.

[0038] Among them, in step A102, the total power consumption time series data is divided into multiple total power consumption time series data segments, and the duration of each total power consumption time series data segment is the length of the time period. The selection of the time period length needs to consider the production characteristics of the enterprise and the purpose of data analysis. For example, for a continuous production enterprise, a shorter time period, such as 1 hour or 2 hours, can be selected; for a batch production enterprise, a time period that matches the production batch can be selected. By reasonably dividing the time period, the periodic changes in the enterprise's power consumption pattern can be better captured.

[0039] Among them, eigendecomposition refers to the process of decomposing a complex signal into multiple simple components, and modal components refer to the basic components of the signal obtained by eigendecomposition. In step A103, eigendecomposition is performed on each total power consumption time series data segment, so that each total power consumption time series data segment can be decomposed into multiple modal components.

[0040] The modal decomposition algorithm can decompose complex signals into multiple simple modal components. Commonly used modal decomposition algorithms include empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD) and variational mode decomposition (VMD). These algorithms have their own characteristics, and the appropriate algorithm can be selected according to actual needs.

[0041] Preferably, in step A103, a variational mode decomposition algorithm is used to perform characteristic decomposition on each total power time series data segment.

[0042] The variational mode decomposition algorithm is an advanced signal processing technology that can effectively decompose complex time series data into multiple modal components. In solving the problem of characteristic decomposition of enterprise total power consumption time series data, the variational mode decomposition algorithm has the following advantages: it can adaptively determine the number of decomposed modes without manual setting; it has strong robustness to noise and can better extract useful signal features; compared with other modal decomposition methods, such as empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD), the variational mode decomposition algorithm can better handle mode aliasing problems.

[0043] The variational modal decomposition algorithm generally achieves the decomposition and extraction of modal components by constructing a constrained variational problem model, introducing Lagrange operators and quadratic penalty factors, iterative optimization, and modal extraction. In this embodiment, the constrained variational problem model can be constructed as: (1); in, is the total power consumption time series data segment, u is the sequence number, is the uth modal component, j is the imaginary number, t is the time, for A collection of is the center frequency of the uth modal component, for A collection of is the pulse function, represents the differential function for the time derivative, and st is the symbol of the constraint condition. By introducing the Lagrange operator and the quadratic penalty factor, and using an iterative algorithm to solve the above constrained variational problem model (the specific process is prior art and will not be described in detail here), each modal component can be obtained. Each modal component obtained based on the above constrained variational problem model has a limited bandwidth, and its frequency band is closely around its specific center frequency. , which can effectively map the unique power characteristics of different electrical equipment and lay a solid foundation for subsequent analysis.

[0044] It should be noted that the constrained variational problem model is not limited to the above-mentioned model, and the constrained variational problem model in the prior art may also be selected.

[0045] The reference power feature set refers to the typical power consumption characteristics of each major production equipment acquired in advance, including time domain characteristics and frequency domain characteristics, which can be obtained by self-reporting by the enterprise or by analyzing the power consumption data of the equipment when it is running alone. The main production equipment is the production equipment that is highly correlated with the business status of the enterprise. The main production equipment can be determined in advance from all the power-consuming equipment of the enterprise manually or through big data statistical methods, and each major production equipment and the corresponding reference power feature set are mapped to form a query table. In step A2, the reference power feature set of each major production equipment of the enterprise is extracted from the query table.

[0046] Among them, time domain features refer to parameters that describe signal characteristics in the time dimension, which can be extracted through statistical analysis methods.

[0047] Among them, frequency domain features refer to parameters that describe signal characteristics in the frequency dimension, which can be extracted through methods such as Fourier transform.

[0048] In some embodiments, the reference time domain feature includes at least one of a reference mean, a reference standard deviation, a reference kurtosis, a reference mean square value, a reference root mean square value, a reference peak factor, a reference pulse factor, and a reference waveform factor; The reference frequency domain feature includes at least one of a reference center frequency, a reference bandwidth, a reference power spectrum density, a reference frequency variance, and a reference center of gravity frequency; The time domain characteristics include at least one of mean, standard deviation, kurtosis, mean square value, root mean square value, peak factor, pulse factor, and waveform factor; The frequency domain characteristics include at least one of center frequency, bandwidth, power spectrum density, frequency variance, and center of gravity frequency.

[0049] Among them, the mean and reference mean reflect the average level of power data; the standard deviation and reference standard deviation characterize the degree of dispersion of power data; the kurtosis and reference kurtosis describe the steepness of the distribution of power data; the mean square value, root mean square value, reference mean square value and reference root mean square value all reflect the energy of power data; the peak factor, pulse factor, waveform factor, reference peak factor, reference pulse factor and reference waveform factor are used to describe the waveform characteristics of power data.

[0050] Among them, the center frequency and reference center frequency indicate the centroid position of the power spectrum; the bandwidth and reference bandwidth reflect the frequency range of the power spectrum; the power spectrum density and reference power spectrum density describe the power distribution within the unit frequency; the frequency variance and reference frequency variance characterize the discrete degree of the frequency component; the centroid frequency and reference centroid frequency reflect the centroid position of the power spectrum.

[0051] The combined use of these features enables a comprehensive description of the characteristics of power data. Through standardized feature definitions, a reliable feature basis is provided for enterprise operating status monitoring. The time domain features reflect the statistical distribution characteristics of power data, and the frequency domain features reflect the frequency composition characteristics of power data. The combined use of the two types of features can comprehensively describe and compare the characteristics of power data. This enables a more accurate matching of the filtered modal components with the main production equipment.

[0052] Specifically, step A3 includes: A301. Calculate the characteristic entropy of each modal component; A302. Compare the characteristic entropy of each modal component with a preset characteristic entropy threshold, and determine the modal component whose characteristic entropy is lower than the characteristic entropy threshold as a valid modal component; A303. Perform noise filtering on the effective modal components to obtain filtered modal components.

[0053] This method effectively solves the problem of modal component validity judgment and noise filtering by calculating characteristic entropy, setting thresholds, judging validity and performing noise filtering. This solution can extract more useful information from complex power data, remove redundancy and noise, and thus improve the accuracy and reliability of subsequent enterprise operating status monitoring.

[0054] Among them, characteristic entropy is an indicator used to measure the complexity of the signal, which can be calculated by using methods such as sample entropy or approximate entropy. The characteristic entropy of each modal component can also be calculated using the following formula: ; ; ; in, is the characteristic entropy, is the first reference quantity, is the second reference quantity, N is the data dimension of the modal component (i.e., the number of data contained in the modal component), M is the preset window length (which can be set according to actual needs), The number of other first data sequences whose distance to the i-th first data sequence is less than or equal to s in the first data sequence for sliding window interception of the modal component using a window of length M with a single data as the step length, s is a preset distance threshold (which can be set according to actual needs), In the second data sequence for sliding window truncation of the modal component using a window with a length of M+1 and a single data as a step, the number of other second data sequences whose distance to the i-th second data sequence is less than or equal to s.

[0055] Among them, the i-th first data sequence can be expressed as: ; is the i-th first data sequence, , , They are the i-th, i+1-th, and i+M-1-th data in the modal component respectively; The distance between any i-th first data sequence and jj-th first data sequence (i≠jj) is: ; is the distance between the i-th first data sequence and the jj-th first data sequence, , They are the i+kth and jj+kth data in the modal components respectively.

[0056] Similarly, the i-th second data sequence can be expressed as: ; is the i-th second data sequence, is the i+Mth data in the modal component.

[0057] The distance between any i-th second data sequence and jj-th second data sequence (i≠jj) is: ; is the distance between the i-th second data sequence and the jj-th second data sequence.

[0058] The size of the characteristic entropy calculated in the above manner reflects the complexity of the power load. The larger the characteristic entropy, the more complex the power load change is and the more difficult it is to predict.

[0059] When the characteristic entropy is too high, it is considered that the corresponding modal component contains too much noise and is not suitable for analyzing the business status of the enterprise. Here, a characteristic entropy threshold is set to judge whether the characteristic entropy of each modal component is too high, so that only the modal components with characteristic entropy lower than the characteristic entropy threshold are used as valid modal components for subsequent business status analysis to ensure the accuracy of the analysis results. Among them, the setting of the characteristic entropy threshold can be adjusted according to the actual application scenario. For example, a suitable threshold range can be determined by statistically analyzing a large amount of sample data.

[0060] Among them, the interference components in the effective modal components can be removed by noise filtering. For the noise filtering of the effective modal components, a variety of filtering algorithms can be used. For example, Wiener filtering, Kalman filtering or wavelet denoising can be used. Furthermore, the most suitable filtering method can be selected according to the characteristics of the signal.

[0061] In some preferred implementations, in step A303, a Wiener adaptive filter algorithm is used to perform noise filtering on effective modal components.

[0062] The Wiener adaptive filter algorithm is an advanced signal processing technology used to extract useful information from noisy signals. For power consumption data, it can effectively remove noise from modal components and improve data quality. Effective modal components are important signal components obtained by characteristic entropy calculation and threshold comparison. Noise filtering of these components can retain key information while reducing unnecessary interference. Noise filtering is designed to improve the signal-to-noise ratio of modal components, making subsequent feature extraction and matching more accurate. By removing noise, the actual power consumption of enterprise production equipment can be more clearly reflected.

[0063] By using the Wiener adaptive filter algorithm to filter the effective modal components, the problem of noise interference in the modal components is effectively solved. This method can adaptively adjust the filter parameters to adapt to the characteristics of different modal components, thereby achieving more accurate noise removal.

[0064] Furthermore, the normalized least mean square error (NLMS) algorithm or the least mean square error (LMS) algorithm can be used to implement the Wiener adaptive filter; preferably, the NLMS algorithm is used to implement the Wiener adaptive filter, which has a faster convergence speed and better stability than the LMS algorithm. The specific methods of using the normalized least mean square error (NLMS) algorithm or the least mean square error (LMS) algorithm to implement the Wiener adaptive filter are all prior arts and will not be described in detail here.

[0065] Further, step A4 includes: A401. Extract the time domain characteristics of each filtered modal component; A402. Extract frequency domain features of each filtered modal component based on FFT transformation; A403. Using the extracted time domain features and frequency domain features to form an actual power feature set of each filtered modal component; A404. Compare the actual power characteristic set of each filtered modal component with the reference power characteristic set of each major production equipment to determine the corresponding relationship between each filtered modal component and each major production equipment.

[0066] By extracting the time domain and frequency domain features of the modal components after filtering and comparing them with the pre-set reference features (i.e., the reference time domain features and reference frequency domain features in the reference power feature set), the corresponding relationship between the modal components after filtering and the main production equipment can be identified. This method not only takes into account the time domain information, but also introduces frequency domain analysis to improve the comprehensiveness and accuracy of feature description. By extracting frequency domain features through FFT transformation, the periodic components in the signal can be captured, which is very helpful for identifying the characteristics of different production equipment. Finally, by comparing the actual power feature set with the reference power feature set, each modal component after filtering can be accurately associated with the corresponding main production equipment. The innovation of this method lies in the comprehensive use of time domain and frequency domain information, which improves the accuracy and reliability of identification and provides an important basis for subsequent enterprise operating status monitoring.

[0067] Among them, the time domain characteristics of the modal components after filtering, such as mean, standard deviation, kurtosis, mean square value, root mean square value, peak factor, pulse factor, waveform factor, etc., can be extracted by statistical analysis methods. This is the existing technology and will not be described in detail here.

[0068] Among them, the frequency domain characteristics of the modal components after filtering, such as center frequency, bandwidth, power spectrum density, frequency variance, centroid frequency, etc., can be extracted by Fourier transform. For example, the modal components after filtering are first converted into corresponding frequency spectra using FFT transform (fast Fourier transform) algorithm, and then the center frequency and bandwidth can be extracted and the power spectrum density can be calculated using the spectrum. Among them, the method for extracting the center frequency and bandwidth and the method for calculating the power spectrum density are prior arts (in fact, when the modal components are extracted using formula (1) in step A1, the center frequency of each modal component can be calculated, so the calculated center frequency can also be directly called here). It will not be described in detail here, and the frequency variance and centroid frequency can be calculated by the following formula: ; ; in, is the frequency variance, is the center of gravity frequency, is the sampling frequency of the total power consumption time series data (for example, when collecting the total power consumption time series data, one data is collected per second, and the sampling frequency is 1 Hz), is the angular frequency, is the power spectrum of the modal component after filtering (it can be calculated based on the frequency spectrum. The calculation method is the existing technology and will not be described in detail here).

[0069] In this embodiment, the working status data includes the start time, working time, stop time and steady-state power value change index; Thus, step A5 includes: A501. Identify the start time, working time and stop time of each major production equipment according to the filtered modal components corresponding to each major production equipment in each time period; A502. Calculate the steady-state power value variation index of each major production equipment based on the characteristic entropy of the filtered modal components corresponding to each major production equipment in each time period.

[0070] The operating status of the enterprise is monitored by analyzing the working status data of the main production equipment of the enterprise. First, the start time, working time and stop time of each main production equipment are identified by using the filtered modal components, which reflects the frequency and intensity of equipment use. Secondly, the stability of equipment operation is evaluated by calculating the steady-state power value change index. This method can more comprehensively and accurately reflect the actual production situation of the enterprise, avoid the misjudgment that may be caused by relying solely on total power consumption, and is also difficult to be manipulated by humans. By analyzing these data, abnormal changes in the operating status of the enterprise can be discovered in a timely manner, providing a more reliable basis for the post-loan management of financial institutions.

[0071] Specifically, the identification of start time, working time and stop time can be achieved through the following steps: First, the filtered modal component is compared with the preset power threshold, and a power threshold is set (which can be set according to actual needs). When the power value exceeds this threshold, the corresponding device is considered to be in the on state; when it is lower than the threshold, the corresponding device is considered to be in the off state.

[0072] Secondly, by analyzing the changing trend of the power value, the start time and stop time of the device can be determined. For example, when the power value suddenly rises from below the power threshold to above the power threshold, this time point can be marked as the start time. Correspondingly, when the power value suddenly drops from above the power threshold to below the power threshold, this time point can be marked as the stop time.

[0073] Again, the working time can be obtained by calculating the time difference between the start time and the stop time.

[0074] Among them, the change of characteristic entropy can reflect the fluctuation of the power load of the corresponding equipment, and the stable characteristic entropy indicates that the power change of the corresponding equipment is relatively stable. Therefore, the fluctuation of the characteristic entropy of the filtered modal component of the same main production equipment in each time period can be used to characterize the change of the steady-state power value of the corresponding main production equipment. For example, the variance or standard deviation of the characteristic entropy of the filtered modal component of the same main production equipment in each time period is calculated as the steady-state power value change index of the corresponding main production equipment. The larger the steady-state power value change index, the more unstable the power load of the corresponding main production equipment.

[0075] Specifically, step A6 includes: A601. Calculate the abnormal value of the operating status based on the start time, working time, stop time and steady-state power value change index; A602. Determine whether the current operating status of the enterprise is abnormal based on the abnormal value of the operating status.

[0076] Among them, the typical start-up time, typical working time, typical stop time and average steady-state power value change index of various main production equipment of enterprises in the same industry as the monitored enterprise under normal business conditions can be obtained in advance. In step A601, the average value (recorded as the first mean value) of the deviation between each start-up time and the typical start-up time, the average value (recorded as the second mean value) of the deviation between each working time and the typical working time, the average value (recorded as the third mean value) of the deviation between each stop time and the typical stop time, and the average value (recorded as the fourth mean value) of the deviation between each steady-state power value change index and the average steady-state power value change index can be calculated for the same main production equipment, and then the weighted sum of the first mean value, the second mean value, the third mean value and the fourth mean value (the weight of each mean value can be adjusted according to actual needs) is calculated as the abnormal index of the corresponding main production equipment, and finally the abnormal value of the business status is calculated according to the abnormal index of each main production equipment, for example, the weighted sum of the abnormal index of each main production equipment (the weight of each main production equipment can be adjusted according to actual needs) is calculated as the abnormal value of the business status, but it is not limited to this.

[0077] When the abnormal value of the operating status is calculated, it is compared with the pre-set abnormal threshold. If the abnormal value exceeds the abnormal threshold, the current operating status of the enterprise is judged to be abnormal; if the abnormal value does not exceed the abnormal threshold, the current operating status of the enterprise is judged to be normal. The abnormal threshold can be determined based on the statistical analysis of historical data. For example, the mean of the abnormal value during the normal operation period plus two standard deviations can be taken as the abnormal threshold.

[0078] In some preferred embodiments, see Figure 2 , after step A6, further comprising the steps of: A7. If the current operating status of the enterprise is normal, the estimated power consumption time series data of the enterprise is predicted based on the total power consumption time series data; the estimated power consumption time series data is the estimated data of the total power consumption time series data of the enterprise in the future; A8. Evaluate the potential risk of abnormal operating status of the enterprise based on the estimated power consumption time series data.

[0079] On the basis that the current business status of the enterprise is normal, the total power consumption time series data is used to predict the estimated power consumption time series data in the future, and based on this, the potential abnormal business status risk is evaluated, thus realizing the prediction and risk assessment of the future business status of the enterprise. This method can not only detect potential risks in a timely manner, but also provide financial institutions with a more forward-looking decision-making basis, effectively improving the efficiency and accuracy of post-loan management.

[0080] There are many methods to predict the time series data of the estimated power consumption of an enterprise. For example, time series analysis methods such as the Autoregressive Integrated Moving Average Model (ARIMA) or the Seasonal Autoregressive Integrated Moving Average Model (SARIMA) can be used. These models can capture the trend, seasonality, and cyclical characteristics of power consumption data, thereby predicting future power consumption.

[0081] Another possible implementation is to use machine learning algorithms such as Long Short-Term Memory (LSTM) or Gated Recurrent Unit (GRU). These deep learning models are particularly suitable for processing time series data and are able to learn long-term dependencies, thereby improving the accuracy of predictions.

[0082] In step A8, the potential abnormal operating status risk of the enterprise is evaluated based on the estimated power consumption time series data. The operating status abnormal value corresponding to the estimated power consumption time series data can be calculated based on the estimated power consumption time series data and steps A1 to A6, and recorded as the estimated operating status abnormal value (specifically, the estimated power consumption time series data is substituted into the total power consumption time series data in steps A1 to A6, and finally the estimated operating status abnormal value is calculated in step A601); and multiple operating status abnormal risk levels and corresponding abnormal value ranges can be set, and the enterprise's operating status abnormal risk level can be determined according to the abnormal value range into which the estimated operating status abnormal value falls.

[0083] In this application, the total power consumption time series data of the enterprise obtained through the previous steps and the current operating status determined provide reliable basic data for predicting future power consumption and assessing potential risks. This synergy makes the entire monitoring and early warning system more comprehensive and accurate, not only able to monitor the current status, but also to predict future risks.

[0084] In summary, the present invention proposes a method for monitoring the business status of an enterprise based on the power data of the enterprise, obtains the time series data of the total power consumption of the enterprise, performs feature decomposition and noise filtering on the data, extracts features and matches them with reference features, calculates the working status data and judges the business status of the enterprise, and predicts the abnormal risks of the potential business status of the enterprise, thereby realizing the in-depth analysis and accurate prediction of the enterprise's power load data. It solves the problems of data formality, information real-time, and automation and digitization of the collection process in the current post-loan management of banks, reduces the workload of managers, and improves the efficiency of supervision; it realizes real-time monitoring and early warning of the business status of the enterprise, which can improve the level of informatization, reduce the workload of loan managers, ensure the authenticity and reliability of data, and improve the real-time and accuracy of data analysis. At the same time, based on the evaluation results, it can also provide enterprises with decision-making support in power management, equipment maintenance, production planning, etc.

[0085] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0086] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for monitoring enterprise operating status based on enterprise power data, characterized in that: Includes steps: A1. Obtain the total power consumption time series data of the enterprise, and perform feature decomposition on the total power consumption time series data according to a preset time period length to obtain multiple modal components corresponding to each time period; A2. Obtain a reference power feature set of each major production equipment of the enterprise; the reference power feature set includes a reference time domain feature and a reference frequency domain feature; A3. Determine the validity of each of the modal components according to the characteristic entropy of the modal components, and perform noise filtering on the valid modal components to obtain filtered modal components; A4. extracting the time domain features and frequency domain features of each of the filtered modal components to form an actual power feature set corresponding to each of the filtered modal components, which is used to match the reference power feature set to identify the corresponding relationship between each of the filtered modal components and each of the main production equipment; A5. Obtain the working status data of each of the main production equipment according to the matching identification results corresponding to each of the time periods; A6. Determine the current operating status of the enterprise based on the working status data.

2. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: Step A1 includes: A101. Obtain the total power consumption time series data of the enterprise for a preset time period; A102. Divide the total power time series data according to the time period length to obtain a plurality of total power time series data segments; A103. Perform feature decomposition on each of the total electric power time series data segments based on a modal decomposition algorithm to obtain multiple modal components corresponding to each of the time periods.

3. The enterprise operation status monitoring method based on enterprise power data according to claim 2 is characterized in that: In step A103, a variational mode decomposition algorithm is used to perform characteristic decomposition on each of the total power time series data segments.

4. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: The reference time domain feature includes at least one of a reference mean, a reference standard deviation, a reference kurtosis, a reference mean square value, a reference root mean square value, a reference peak factor, a reference pulse factor, and a reference waveform factor; The reference frequency domain feature includes at least one of a reference center frequency, a reference bandwidth, a reference power spectrum density, a reference frequency variance, and a reference center of gravity frequency; The time domain characteristics include at least one of mean, standard deviation, kurtosis, mean square value, root mean square value, peak factor, pulse factor, and waveform factor; The frequency domain characteristics include at least one of center frequency, bandwidth, power spectrum density, frequency variance, and center of gravity frequency.

5. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: Step A3 includes: A301. Calculate the characteristic entropy of each of the modal components; A302. Compare the characteristic entropy of each of the modal components with a preset characteristic entropy threshold, and determine the modal component whose characteristic entropy is lower than the characteristic entropy threshold as a valid modal component; A303. Perform noise filtering on the effective modal components to obtain filtered modal components.

6. The enterprise operation status monitoring method based on enterprise power data according to claim 5 is characterized in that: In step A303, a Wiener adaptive filter algorithm is used to perform noise filtering processing on the effective modal components.

7. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: Step A4 includes: A401. Extracting the time domain features of each of the filtered modal components; A402. Extracting the frequency domain characteristics of each of the filtered modal components based on FFT transform; A403. Using the extracted time domain features and the frequency domain features to form an actual power feature set of each of the filtered modal components; A404. Compare the actual power characteristic set of each of the filtered modal components with the reference power characteristic set of each of the main production equipment, and determine the corresponding relationship between each of the filtered modal components and each of the main production equipment.

8. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: The working status data includes the start time, working time, stop time and steady-state power value change index; Step A5 includes: According to the filtered modal components corresponding to each of the main production equipment in each of the time periods, identifying the start time, working time and stop time of each of the main production equipment; According to the characteristic entropy of the filtered modal component corresponding to each of the main production equipment in each of the time periods, the steady-state power value variation index of each of the main production equipment is calculated.

9. The enterprise operation status monitoring method based on enterprise power data according to claim 8 is characterized in that: Step A6 includes: Calculate an abnormal value of the operating state according to the start time, the working time, the stop time and the steady-state power value change index; Whether the current operating status of the enterprise is abnormal is determined according to the operating status abnormal value.

10. The enterprise operation status monitoring method based on enterprise power data according to claim 1 is characterized in that: After step A6, the method further includes the following steps: A7. If the current operating status of the enterprise is normal, predict the estimated power consumption time series data of the enterprise based on the total power consumption time series data; the estimated power consumption time series data is the estimated data of the total power consumption time series data of the enterprise in the future; A8. Evaluate the potential risk of abnormal operating status of the enterprise based on the estimated power consumption time series data.

Images