Industrial hydropower energy consumption abnormity monitoring method and system

Abstract

The invention provides an industrial hydropower energy consumption abnormity monitoring method and system, and the method comprises the steps: periodically obtaining industrial water consumption and industrial electricity consumption, and obtaining a water energy sequence and an electric energy sequence; based on the water energy sequence and the electric energy sequence, determining corresponding water energy characteristics and electric energy characteristics; according to the water energy characteristics and the electric energy characteristics, constructing water and electricity associated characteristics; processing the water energy sequence and the electric energy sequence according to a pre-trained space-time prediction model to obtain predicted water energy and predicted electric energy under the condition that it is determined that the water energy sequence and the electric energy sequence are not abnormal according to the water and electricity association characteristics; and outputting abnormal monitoring results of the water energy sequence and the electric energy sequence based on the predicted water energy and the predicted electric energy. The invention relates to the technical field of industrial monitoring, and can improve the accuracy of monitoring the energy consumption abnormity of industrial hydropower.

Description

Technical Field

[0001] This application relates to the field of industrial monitoring technologies, and particularly to an industrial water and electricity energy consumption anomaly monitoring method and system. Background Art

[0002] Currently, with the development of data science, some enterprises or manufacturers tend to use big data technologies to empower industrial production in order to improve the efficiency of industrial production and reduce the cost of industrial production. For example, some enterprises monitor water and electricity energy consumption to optimize the process, thereby improving the benefits of industrial production. Currently, industrial process technologies are usually adjusted based on the energy consumption data collected in real time by sensors. However, this method does not consider deep data features such as the correlation of water and electricity energy consumption, resulting in low accuracy of industrial monitoring. Summary of the Invention

[0003] In view of the above, it is necessary to propose an industrial water and electricity energy consumption anomaly monitoring method and system to solve the technical problem of low accuracy of industrial monitoring.

[0004] This application provides an industrial water and electricity energy consumption anomaly monitoring method, which is applied to an electronic device. The method includes: periodically obtaining industrial water energy consumption and industrial electricity energy consumption to obtain a water energy sequence and an electricity energy sequence; determining corresponding water energy characteristics and electricity energy characteristics based on the water energy sequence and the electricity energy sequence; constructing a water and electricity correlation feature according to the water energy characteristics and the electricity energy characteristics; in the case where it is determined that the water energy sequence and the electricity energy sequence are not abnormal according to the water and electricity correlation feature, processing the water energy sequence and the electricity energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electricity energy; outputting an anomaly monitoring result of the water energy sequence and the electricity energy sequence based on the predicted water energy and the predicted electricity energy.

[0005] In some embodiments, constructing the water and electricity correlation feature according to the water energy characteristics and the electricity energy characteristics includes: determining a water and electricity coupling coefficient of the water energy characteristics and the electricity energy characteristics according to the power data in the electricity energy characteristics, the flow data and the pressure difference data in the water energy characteristics; determining an energy consumption matching degree of the water energy characteristics and the electricity energy characteristics according to the first actual energy consumption and the first planned energy consumption in the water energy characteristics and the second actual energy consumption and the second planned energy consumption in the electricity energy characteristics; determining peak-time energy consumption and valley-time energy consumption from the second actual energy consumption according to the time stamp of the electricity energy characteristics; determining an energy consumption distribution coefficient of the electricity energy characteristics according to the peak-time energy consumption and the valley-time energy consumption; determining an indirect correlation coefficient of the water energy characteristics and the electricity energy characteristics according to the energy consumption distribution coefficient and the energy consumption matching degree; determining the water and electricity correlation feature according to the water and electricity coupling coefficient and the indirect correlation coefficient.

[0006] In some embodiments, determining the hydro - electric coupling coefficient of the water energy feature and the electric energy feature includes: determining the product of the flow rate data and the pressure difference data; determining the hydro - electric coupling coefficient according to the ratio of the power data to the product; wherein, the hydro - electric coupling coefficient is used to indicate the degree of association between the water energy feature and the electric energy feature.

[0007] In some embodiments, determining the indirect correlation coefficient of the water energy feature and the electric energy feature according to the energy consumption distribution coefficient and the energy consumption matching degree includes: determining the product of the energy consumption distribution coefficient and the energy consumption matching degree as the indirect correlation coefficient of the water energy feature and the electric energy feature.

[0008] In some embodiments, determining that there are anomalies in the water energy sequence and the electric energy sequence according to the hydro - electric correlation feature includes: constructing nodes according to the water energy feature and the electric energy feature; determining the edge weights between any two nodes according to the hydro - electric coupling coefficient and the indirect correlation coefficient; constructing hydro - electric graph data according to the nodes and the edge weights; determining the encoded vector of the hydro - electric graph data based on a pre - trained graph neural network; determining that there are anomalies in the water energy sequence and the electric energy sequence when the difference degree between the encoded vector and a pre - stored historical encoded vector is greater than or equal to a preset difference threshold; determining that there are no anomalies in the water energy sequence and the electric energy sequence when the difference degree between the encoded vector and a pre - stored historical encoded vector is less than the preset difference threshold.

[0009] In some embodiments, the method further includes training the spatio - temporal prediction model, and training the spatio - temporal prediction model includes: constructing a first time - domain encoding and a second time - domain encoding according to pre - obtained historical water energy sequences, historical electric energy sequences, and corresponding timestamps; constructing a first spatial - domain encoding and a second spatial - domain encoding according to the historical water energy sequences, the historical electric energy sequences, and corresponding position information; determining corresponding predicted time - domain encoding and predicted spatial - domain encoding based on the first time - domain encoding and the first spatial - domain encoding according to a pre - constructed initial prediction model; determining a first loss value of the initial prediction model according to the difference between the second time - domain encoding and the predicted time - domain encoding, and the difference between the second spatial - domain encoding and the predicted spatial - domain encoding; updating the initial prediction model according to the back - propagation algorithm until the first loss value meets a preset condition, and stopping updating the initial prediction model to obtain a spatio - temporal prediction model trained to a convergent state.

[0010] In some embodiments, the method for determining the first loss value of the initial prediction model satisfies the following first loss function: ; Wherein, Loss represents the first loss value; α represents the first weight parameter of the difference between the second time-domain encoding and the predicted time-domain encoding; β represents the second weight parameter of the difference between the second spatial-domain encoding and the predicted spatial-domain encoding; γ represents the third weight parameter of the similarity between the second time-domain encoding and the second spatial-domain encoding; i represents the index of the dimension in the second time-domain encoding, and n represents the number of dimensions in the second time-domain encoding; j represents the index of the dimension in the second spatial-domain encoding, and m represents the number of dimensions in the second spatial-domain encoding; A_i represents the value of the i-th dimension in the second time-domain encoding; B_i represents the value of the i-th dimension in the predicted time-domain encoding; C_j represents the value of the j-th dimension in the second spatial-domain encoding; D_j represents the value of the j-th dimension in the predicted spatial-domain encoding; G represents the similarity between the second time-domain encoding and the second spatial-domain encoding.

[0011] In some embodiments, the method further includes: when the number of times of determining the predicted water energy and the predicted electric energy based on the spatio-temporal prediction model reaches a preset number threshold, determining the water energy sequence and the electric energy sequence corresponding to the predicted water energy and the predicted electric energy output by the spatio-temporal prediction model for the last time; and updating the first loss function based on the similarity between the water energy sequence and the electric energy sequence.

[0012] An embodiment of the present application further provides an industrial hydropower energy consumption anomaly monitoring system, where the system includes an electronic device and a sensor; the electronic device is configured to periodically obtain industrial water energy consumption and industrial electricity energy consumption from the sensor to obtain a water energy sequence and an electric energy sequence; the electronic device is further configured to determine corresponding water energy characteristics and electric energy characteristics based on the water energy sequence and the electric energy sequence; the electronic device is further configured to construct a hydropower correlation characteristic according to the water energy characteristics and the electric energy characteristics; the electronic device is further configured to, when it is determined that there is no anomaly in the water energy sequence and the electric energy sequence according to the hydropower correlation characteristic, process the water energy sequence and the electric energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electric energy; the electronic device is further configured to output an anomaly monitoring result of the water energy sequence and the electric energy sequence based on the predicted water energy and the predicted electric energy.

[0013] In some embodiments, the electronic device constructs a hydropower correlation feature based on the water energy feature and the electric energy feature, including: determining a hydropower coupling coefficient of the water energy feature and the electric energy feature according to the power data in the electric energy feature, the flow rate data and the pressure difference data in the water energy feature; determining an energy consumption matching degree of the water energy feature and the electric energy feature according to the first actual energy consumption and the first planned energy consumption in the water energy feature and the second actual energy consumption and the second planned energy consumption in the electric energy feature; determining peak-hour energy consumption and valley-hour energy consumption from the second actual energy consumption according to the time stamp of the electric energy feature; determining an energy consumption distribution coefficient of the electric energy feature according to the peak-hour energy consumption and the valley-hour energy consumption; determining an indirect correlation coefficient of the water energy feature and the electric energy feature according to the energy consumption distribution coefficient and the energy consumption matching degree; and determining the hydropower correlation feature according to the hydropower coupling coefficient and the indirect correlation coefficient As can be seen from the above technical solutions, in the embodiments of the present application, by periodically acquiring industrial water and electricity energy consumption data, a water energy sequence and an electric energy sequence are constructed, and the data is refined to the minute level to improve the spatio-temporal resolution of monitoring. Based on the sequences, water energy features and electric energy features (such as mean value, variance, trend slope, etc.) are extracted, and hydropower correlation features (such as energy consumption ratio, phase difference) are constructed to capture subtle changes in the energy consumption pattern. Using a pre-trained spatio-temporal prediction model, combining historical data with real-time data, the energy consumption value in a future period is predicted. When the deviation between the predicted value and the actual value exceeds the threshold, an abnormal warning is triggered. Based on the predicted water energy and electric energy, the equipment start-stop strategy is optimized. The production process is adjusted in combination with the energy consumption data, real-time abnormal events are provided to realize automatic abnormal processing, and the model is continuously trained with new data to improve the prediction accuracy Description of the Drawings

[0014] Figure 1 FIG. is an application scenario diagram of an industrial hydropower energy consumption abnormal monitoring method provided by an embodiment of the present application

[0015] Figure 2 FIG. is a flowchart of an industrial hydropower energy consumption abnormal monitoring method provided by an embodiment of the present application

[0016] Figure 3 FIG. is a functional module diagram of an industrial hydropower energy consumption abnormal monitoring system provided by an embodiment of the present application

[0017] Figure 4 FIG. is a structural schematic diagram of an electronic device provided by an embodiment of the present application Detailed Embodiments

[0018] In order to more clearly understand the purpose, features, and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other. Many specific details are set forth in the following description in order to fully understand the present application. The described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0019] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present application, "a plurality" means two or more, unless otherwise specifically defined.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0021] The embodiments of the present application provide an industrial water and electricity energy consumption anomaly monitoring method, which can be applied to one or more electronic devices. An electronic device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, a microprocessor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0022] An electronic device can be any electronic product that can perform human-computer interaction with a customer. For example, a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an Internet Protocol Television (IPTV), a smart wearable device, etc.

[0023] The electronic device may further include a network device and / or a client device. Among them, the network device includes, but is not limited to, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on Cloud Computing.

[0024] The network where the electronic device is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, Virtual Private Network (VPN), etc.

[0025] As Figure 1 shown is an application scenario diagram of an industrial water and electricity energy consumption anomaly monitoring method provided by an embodiment of the present application. An industrial water and electricity energy consumption anomaly monitoring method provided by the present application can be applied to the electronic device 100. Among them, the electronic device 100 can be a device such as a computer or a server applied to a monitoring site, and the electronic device 100 is communicatively connected to the sensor 200. The electronic device 100 is configured to periodically obtain industrial water energy consumption and industrial electricity energy consumption from the sensor 200 to obtain a water energy sequence and an electric energy sequence. The electronic device 100 is further configured to determine corresponding water energy characteristics and electric energy characteristics based on the water energy sequence and the electric energy sequence. The electronic device 100 is further configured to construct a water and electricity correlation feature according to the water energy characteristics and the electric energy characteristics. The electronic device 100 is further configured to, when it is determined that the water energy sequence and the electric energy sequence are normal according to the water and electricity correlation feature, process the water energy sequence and the electric energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electric energy. The electronic device 100 is further configured to output an anomaly monitoring result of the water energy sequence and the electric energy sequence based on the predicted water energy and the predicted electric energy.

[0026] As Figure 2 shown is a flowchart of an industrial water and electricity energy consumption anomaly monitoring method provided by an embodiment of the present application. According to different requirements, the order of the steps in this flowchart can be changed, and some steps can be omitted. An industrial water and electricity energy consumption anomaly monitoring method provided by an embodiment of the present application includes the following steps.

[0027] S20, periodically obtain industrial water energy consumption and industrial electricity energy consumption to obtain a water energy sequence and an electric energy sequence.

[0028] In an embodiment of the present application, during the process of monitoring the energy consumption of industrial water and industrial electricity, the industrial water energy consumption and the industrial electricity energy consumption can be periodically obtained first to provide data support for predicting the industrial water energy consumption and the industrial electricity energy consumption in any time period in the future.

[0029] Specifically, industrial water consumption energy refers to the water resources consumed and related energy consumption data during the production process of industrial enterprises, including: the total water volume used for industrial production activities by the enterprise during the reporting period, including the fresh water intake volume and the recycled water volume; water consumption by different departments, such as main production water (directly used in the product production process), auxiliary production water (such as for support departments like machine repair, transportation, and air compressor stations), and affiliated production water (such as for in-plant greening, staff canteens, and bathrooms); water consumption peaks and valleys, to record the water consumption fluctuations in different time periods (such as daily, monthly, quarterly) for analyzing water consumption patterns; water intake methods: distinguishing between water directly extracted from water sources or purchased from the market (such as reclaimed water, mineral water, etc.); external water supply volume: the water volume supplied by the enterprise to external units (such as raw water, tap water sold, etc.), based on the off-site water volume; external drainage volume: the water volume discharged to the outside after being used in the production process, including the water volume entering the sewage pipe network or the natural environment.

[0030] In an embodiment of the present application, the industrial water consumption energy data further includes: the industrial water reuse rate, which is used to indicate the percentage of the recycled water volume in the total water volume and reflects the recycling efficiency of water resources; the water consumption per unit product, which is used to indicate the water volume consumed for producing each unit of product and is used to evaluate the water-saving level in the production process.

[0031] In an embodiment of the present application, industrial electricity consumption energy refers to the electrical energy consumed and related cost data during the production process of industrial enterprises, including: the total electricity consumption, which is used to indicate the total electricity volume consumed by the enterprise during the reporting period, including the electricity volumes in different time periods such as peaks, super peaks, flats, and valleys; electricity consumption by different equipment / areas, which is used to record the electricity consumption of key equipment or production areas for identifying high-energy-consuming links; voltage and current, which are used to monitor the operating status of equipment to ensure power supply stability; power factor, which is used to reflect the efficiency of electrical equipment, and too low a power factor may lead to an increase in the power adjustment electricity fee. In an embodiment of the present application, the water energy sequence is the industrial water consumption energy data recorded in chronological order and is used to analyze and predict water consumption trends, including: timestamp, which is used to record the specific time of data collection (such as daily, monthly, annually); water consumption data, which is used to indicate the total water consumption, water consumption by different departments, external water supply volume, external drainage volume, etc.; water quality parameters, which are used to indicate water intake methods, external water supply volume, external drainage volume, etc.; water use efficiency indicators, which are used to indicate the industrial water reuse rate, water consumption per unit product, etc.

[0032] In an embodiment of the present application, the electrical energy sequence is the industrial electricity consumption energy data recorded in chronological order and is used to analyze and predict electricity consumption trends, including: timestamp, which is used to record the specific time of data collection (such as hourly, daily, monthly); electricity consumption data, which is used to indicate the total electricity consumption, electricity consumption by different equipment / areas, electricity consumption in peak and valley periods, etc.; electrical energy quality parameters, which are used to indicate voltage, current, power factor, etc.

[0033] S21. Based on the water energy sequence and the electric energy sequence, determine the corresponding water energy characteristics and electric energy characteristics.

[0034] In an embodiment of the present application, the water energy characteristic is an index describing the energy consumption pattern of industrial water use, and the electric energy characteristic is an index describing the energy consumption pattern of industrial electricity use. Exemplarily, the water energy characteristics include statistical characteristics, time domain characteristics, frequency domain characteristics, model fitting characteristics, complexity metric characteristics, etc. Among them, the statistical characteristics include: mean / median, which is used to reflect the average level or typical value of water consumption and electricity consumption; variance / standard deviation, which is used to measure the fluctuation degree of water consumption and electricity consumption; maximum / minimum value, which is used to identify the peaks and valleys of water or electricity use; skewness / kurtosis, which is used to describe the symmetry and sharpness of the distribution of water consumption or electricity consumption. The time domain characteristics include: peak value / trough value, which is used to indicate the instantaneous mutation of water consumption or electricity consumption (such as equipment start-stop); zero crossing point, which is used to analyze the change of water flow direction (such as the alternation of positive and negative flow rates); volatility, which is used to quantify the short-term change rate of water consumption or electricity consumption. The frequency domain characteristics include: periodic components, which are used to identify daily / weekly / monthly periodic fluctuations through Fourier transform; spectral entropy, which is used to evaluate the randomness of water use patterns or electricity use patterns (a high entropy value indicates a complex pattern). The model fitting characteristics include: ARIMA model parameters, which are used to indicate the autoregressive and moving average characteristics of the water consumption sequence or electricity consumption sequence; seasonal intensity, which is used to quantify the influence of seasonal factors on water consumption or electricity consumption. The complexity metric characteristics include: fractal dimension, which is used to describe the roughness and self-similarity of the water consumption sequence or electricity consumption sequence; sample entropy, which is used to measure the predictability of water use patterns or electricity use patterns (a low entropy value indicates high regularity).

[0035] In an embodiment of the present application, the electric energy characteristics are used to characterize the load characteristics and power quality of industrial electricity use. Exemplarily, the electric energy characteristics include: statistical characteristics, time domain characteristics, frequency domain characteristics, model fitting characteristics, etc. Among them, the statistical characteristics include: total electricity consumption / peak-valley difference, which reflects the overall energy consumption and load fluctuation; power factor, which is used to evaluate the electricity use efficiency (a low power factor leads to an increase in the power adjustment electricity bill); effective value of voltage / current, which is used to monitor the stability of power supply. The time domain characteristics include: harmonic content, which is used to analyze the harmonic components in voltage / current through wavelet transform algorithm; voltage sag / surge, which is used to capture instantaneous power quality problems (such as equipment failures); three-phase unbalance degree, which is used to evaluate the symmetry of the power system. The frequency domain characteristics include: power spectral density, which is used to identify the energy distribution at specific frequencies (such as motor harmonics); frequency band energy ratio, which is used to analyze the energy proportion of signals in different frequency bands. The model fitting characteristics include: LSTM hidden state, which is used to capture the long-term dependence relationship of electricity consumption; prediction error distribution, which is used to reflect the fitting ability of the model to electricity use patterns.

[0036] S22. Construct the hydropower correlation characteristics according to the water energy characteristics and the electric energy characteristics.

[0037] In an embodiment of the present application, constructing the hydropower correlation feature according to the water energy feature and the electric energy feature includes: determining the hydropower coupling coefficient of the water energy feature and the electric energy feature according to the power data in the electric energy feature, the flow data and the pressure difference data in the water energy feature; determining the energy consumption matching degree of the water energy feature and the electric energy feature according to the first actual energy consumption and the first planned energy consumption in the water energy feature and the second actual energy consumption and the second planned energy consumption in the electric energy feature; determining the peak-hour energy consumption and the valley-hour energy consumption from the second actual energy consumption according to the timestamp of the electric energy feature; determining the energy consumption distribution coefficient of the electric energy feature according to the peak-hour energy consumption and the valley-hour energy consumption; determining the indirect correlation coefficient of the water energy feature and the electric energy feature according to the energy consumption distribution coefficient and the energy consumption matching degree; and determining the hydropower correlation feature according to the hydropower coupling coefficient and the indirect correlation coefficient.

[0038] In an embodiment of the present application, determining the hydropower coupling coefficient of the water energy feature and the electric energy feature includes: determining the product of the flow data and the pressure difference data; determining the hydropower coupling coefficient according to the ratio of the power data to the product; wherein the hydropower coupling coefficient is used to indicate the degree of association between the water energy feature and the electric energy feature.

[0039] In an embodiment of the present application, determining the indirect correlation coefficient of the water energy feature and the electric energy feature according to the energy consumption distribution coefficient and the energy consumption matching degree includes: determining the product of the energy consumption distribution coefficient and the energy consumption matching degree as the indirect correlation coefficient of the water energy feature and the electric energy feature.

[0040] S23. When it is determined that the water energy sequence and the electric energy sequence are normal according to the hydropower correlation feature, processing the water energy sequence and the electric energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electric energy.

[0041] In an embodiment of the present application, determining that the water energy sequence and the electric energy sequence are abnormal according to the hydropower correlation feature includes: constructing nodes according to the water energy feature and the electric energy feature; determining the edge weights between any two nodes according to the hydropower coupling coefficient and the indirect correlation coefficient; constructing hydropower graph data according to the nodes and the edge weights; determining the encoded vector of the hydropower graph data based on a pre-trained graph neural network; determining that the water energy sequence and the electric energy sequence are abnormal when the difference degree between the encoded vector and a pre-stored historical encoded vector is greater than or equal to a preset difference threshold; and determining that the water energy sequence and the electric energy sequence are normal when the difference degree between the encoded vector and the pre-stored historical encoded vector is less than the preset difference threshold.

[0042] In an embodiment of the present application, the method further includes training the spatio-temporal prediction model, and the training of the spatio-temporal prediction model includes: constructing a first time-domain encoding and a second time-domain encoding according to a pre-acquired historical water energy sequence, a historical electric energy sequence, and corresponding timestamps; constructing a first spatial-domain encoding and a second spatial-domain encoding according to the historical water energy sequence, the historical electric energy sequence, and corresponding position information; determining corresponding predicted time-domain encoding and predicted spatial-domain encoding based on the first time-domain encoding and the first spatial-domain encoding according to an initially constructed prediction model; determining a first loss value of the initial prediction model according to the difference between the second time-domain encoding and the predicted time-domain encoding, and the difference between the second spatial-domain encoding and the predicted spatial-domain encoding; updating the initial prediction model according to the backpropagation algorithm until the first loss value meets a preset condition, and stopping updating the initial prediction model to obtain a spatio-temporal prediction model trained to a convergent state.

[0043] In an embodiment of the present application, the method for determining the first loss value of the initial prediction model satisfies the following first loss function: ; where Loss represents the first loss value; represents a first weight parameter for the difference between the second time-domain encoding and the predicted time-domain encoding; represents a second weight parameter for the difference between the second spatial-domain encoding and the predicted spatial-domain encoding; represents a third weight parameter for the similarity between the second time-domain encoding and the second spatial-domain encoding; i represents the index of the dimension in the second time-domain encoding, and n represents the number of dimensions in the second time-domain encoding; j represents the index of the dimension in the second spatial-domain encoding, and m represents the number of dimensions in the second spatial-domain encoding; represents the value of the i-th dimension in the second time-domain encoding; represents the value of the i-th dimension in the predicted time-domain encoding; represents the value of the j-th dimension in the second spatial-domain encoding; represents the value of the j-th dimension in the predicted spatial-domain encoding; G represents the similarity between the second time-domain encoding and the second spatial-domain encoding. When the first loss value is smaller, it indicates that the difference between the predicted time-domain encoding output by the initial prediction model and the second time-domain encoding is smaller, and it also indicates that the difference between the predicted spatial-domain encodings output by the initial prediction model is smaller. Then, the accuracy of the prediction result output by the initial prediction model is higher.

[0044] In an embodiment of the present application, the method further includes: when the number of times of determining the predicted water energy and the predicted electric energy based on the spatio-temporal prediction model reaches a preset number threshold, determining the water energy sequence and the electric energy sequence corresponding to the predicted water energy and the predicted electric energy output by the spatio-temporal prediction model for the last time; and updating the first loss function based on the similarity between the water energy sequence and the electric energy sequence.

[0045] S24. Output the anomaly monitoring results of the water energy sequence and the electric energy sequence based on the predicted water energy and the predicted electric energy.

[0046] In an embodiment of the present application, the anomalies of the water energy sequence and the electric energy sequence can be classified according to the predicted water energy and the predicted electric energy to obtain the anomaly monitoring results of the water energy sequence and the electric energy sequence. Exemplarily, when the predicted water energy or the predicted electric energy exceeds a preset threshold, or the ratio of the predicted water energy to the predicted electric energy exceeds a ratio threshold, it indicates that a single index (for example, the predicted water energy or the predicted electric energy) exceeds the expectation, or the coupling degree between multiple indexes is abnormal. Therefore, the anomaly level of the anomaly monitoring result can be determined to be relatively low; when multiple predicted water energies or multiple predicted electric energies obtained continuously for multiple times exceed the preset threshold, or the ratio of multiple predicted water energies to multiple predicted electric energies exceeds the ratio threshold, it indicates that there are persistent anomalies in the time domain dimension of industrial water consumption energy and industrial electricity consumption energy. Then, the anomaly level of the anomaly monitoring result can be determined to be relatively high.

[0047] It can be seen from the above technical solutions that in the embodiment of the present application, by periodically acquiring industrial water and electricity consumption data, constructing a water energy sequence and an electric energy sequence, refining the data to the minute level, the spatio-temporal resolution of monitoring is improved. Water energy features and electric energy features (such as mean value, variance, trend slope, etc.) are extracted based on the sequence, and hydroelectric association features (such as energy consumption ratio, phase difference) are constructed to capture subtle changes in the energy consumption pattern. Using a pre-trained spatio-temporal prediction model, combining historical data with real-time data to predict the energy consumption value in the future period. When the deviation between the predicted value and the actual value exceeds the threshold, an anomaly warning is triggered. Based on the predicted water energy and electric energy, the equipment start-stop strategy is optimized. The production process is adjusted in combination with the energy consumption data, real-time anomaly events are provided to realize automatic anomaly processing, and the model is continuously trained with new data to improve the prediction accuracy.

[0048] Please refer to Figure 3 , Figure 3 which is a functional module diagram of an industrial water and electricity energy consumption anomaly monitoring system provided by an embodiment of the present application. The industrial water and electricity energy consumption anomaly monitoring system 31 includes an electronic device 100 and a sensor 200. The module / unit referred to in the present application means a series of computer-readable instruction segments that can be executed by a processor 13 and can complete a fixed function, and is stored in a memory 12. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0049] The electronic device 100 is configured to periodically obtain the industrial water energy consumption and industrial electricity energy consumption from the sensor 200, and obtain a water energy sequence and an electricity energy sequence.

[0050] The electronic device 100 is further configured to determine corresponding water energy characteristics and electricity energy characteristics based on the water energy sequence and the electricity energy sequence.

[0051] The electronic device 100 is further configured to construct a water-electricity correlation feature according to the water energy characteristics and the electricity energy characteristics.

[0052] The electronic device 100 is further configured to, when it is determined that the water energy sequence and the electricity energy sequence are normal according to the water-electricity correlation feature, process the water energy sequence and the electricity energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electricity energy.

[0053] The electronic device 100 is further configured to output an anomaly monitoring result of the water energy sequence and the electricity energy sequence based on the predicted water energy and the predicted electricity energy.

[0054] In some embodiments, the electronic device 100 constructs a water-electricity correlation feature according to the water energy characteristics and the electricity energy characteristics, including: determining a water-electricity coupling coefficient of the water energy characteristics and the electricity energy characteristics according to the power data in the electricity energy characteristics, the flow data and the pressure difference data in the water energy characteristics; determining an energy consumption matching degree of the water energy characteristics and the electricity energy characteristics according to the first actual energy consumption and the first planned energy consumption in the water energy characteristics and the second actual energy consumption and the second planned energy consumption in the electricity energy characteristics; determining peak-time energy consumption and valley-time energy consumption from the second actual energy consumption according to the time stamp of the electricity energy characteristics; determining an energy consumption distribution coefficient of the electricity energy characteristics according to the peak-time energy consumption and the valley-time energy consumption; determining an indirect correlation coefficient of the water energy characteristics and the electricity energy characteristics according to the energy consumption distribution coefficient and the energy consumption matching degree; and determining the water-electricity correlation feature according to the water-electricity coupling coefficient and the indirect correlation coefficient.

[0055] Please refer to Figure 4 , which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 100 includes a memory 12 and a processor 13. The memory 12 is used to store computer-readable instructions, and the processor 13 is configured to execute the computer-readable instructions stored in the memory to implement an industrial water and electricity energy consumption anomaly monitoring method according to any one of the above embodiments.

[0056] In an embodiment of the present application, the electronic device 100 further includes a bus and a computer program stored in the memory 12 and executable on the processor 13, such as an industrial water and electricity energy consumption anomaly monitoring program.

[0057] Figure 4 Only the electronic device 100 with a memory 12 and a processor 13 is shown. Those skilled in the art can understand that Figure 3 the shown structure does not constitute a limitation on the electronic device 100, and it may include fewer or more components than shown, or combine certain components, or have a different component arrangement.

[0058] In combination with Figure 2 , the memory 12 in the electronic device 100 stores a plurality of computer-readable instructions to implement the industrial water and electricity energy consumption anomaly monitoring method. The processor 13 can execute the plurality of instructions to thereby implement: periodically obtaining the industrial water energy consumption and the industrial electricity energy consumption to obtain a water energy sequence and an electricity energy sequence; determining corresponding water energy characteristics and electricity energy characteristics based on the water energy sequence and the electricity energy sequence; constructing a water and electricity correlation characteristic according to the water energy characteristics and the electricity energy characteristics; in the case where it is determined according to the water and electricity correlation characteristic that the water energy sequence and the electricity energy sequence are normal, processing the water energy sequence and the electricity energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electricity energy; and outputting an anomaly monitoring result of the water energy sequence and the electricity energy sequence based on the predicted water energy and the predicted electricity energy.

[0059] Specifically, for the specific implementation method of the above instructions by the processor 13, reference may be made to Figure 2 the description of the relevant steps in the corresponding embodiment, which will not be elaborated here.

[0060] Those skilled in the art can understand that the schematic diagram is only an example of the electronic device 100 and does not constitute a limitation on the electronic device 100. The electronic device 100 may be a bus structure or a star structure. The electronic device 100 may also include more or fewer other hardware or software than shown, or a different component arrangement. For example, the electronic device 100 may also include input / output devices, network access devices, etc.

[0061] It should be noted that the electronic device 100 is only an example. Other existing or future possible electronic products that can be adapted to this application should also be included within the protection scope of this application and are included herein by reference.

[0062] Among them, the memory 12 includes at least one type of readable storage medium, which can be non-volatile or volatile. The readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 12 can be an internal storage unit of the electronic device 100 in some embodiments, such as the mobile hard disk of the electronic device 100. The memory 12 can also be an external storage device of the electronic device 100 in some other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a FlashCard, etc. equipped on the electronic device 100. The memory 12 can be used not only to store application software installed on the electronic device 100 and various types of data, such as the code of an industrial water and electricity energy consumption anomaly monitoring program, etc., but also to temporarily store data that has been output or will be output.

[0063] The processor 13 can be composed of integrated circuits in some embodiments. For example, it can be composed of a single packaged integrated circuit, or can be composed of multiple integrated circuits with the same or different functions packaged together, including a combination of one or more Central Processing Units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 13 is the control core (Control Unit) of the electronic device 100, connecting various components of the entire electronic device 100 through various interfaces and lines, and by running or executing programs or modules stored in the memory 12 (such as executing an industrial water and electricity energy consumption anomaly monitoring program, etc.), and calling data stored in the memory 12, to execute various functions of the electronic device 100 and process data.

[0064] The processor 13 executes the operating system of the electronic device 100 and various installed application programs. The processor 13 executes the application program to implement the steps in the above-mentioned embodiments of various industrial water and electricity energy consumption anomaly monitoring methods, such as Figure 2 the steps shown.

[0065] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory 12 and executed by the processor 13 to complete this application. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device 100.

[0066] The integrated units implemented in the form of software functional modules can be stored in a computer-readable storage medium. The above-mentioned software functional modules are stored in a storage medium and include several instructions for causing a computer device (which may be a personal computer, a computer device, or a network device, etc.) or a processor to execute a part of the industrial water and electricity energy consumption anomaly monitoring method described in various embodiments of the present application.

[0067] If the integrated module / unit of the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present application, it can also be completed by a computer program instructing relevant hardware devices. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented.

[0068] Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory, and other memories, etc.

[0069] Furthermore, the computer-readable storage medium mainly includes a storage program area and a storage data area. Among them, the storage program area can store an operating system, application programs required for at least one function, etc.; the storage data area can store data created according to the use of blockchain nodes, etc.

[0070] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, in Figure 4 only one arrow is used to represent it, but it does not mean that there is only one bus or one type of bus. The bus is set to realize the connection and communication between the memory 12 and at least one processor 13, etc.

[0071] The embodiments of the present application also provide a computer-readable storage medium (not shown in the figure). Computer-readable instructions are stored in the computer-readable storage medium and are executed by a processor in an electronic device to implement the industrial water and electricity energy consumption anomaly monitoring method described in any of the above embodiments.

[0072] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0073] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0074] In addition, in each embodiment of the present application, the functional modules can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a combination of hardware and software functional modules.

[0075] In addition, obviously, the term "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices described in the specification can also be implemented by one unit or device through software or hardware. Terms such as first and second are used to represent names and do not indicate any specific order.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An abnormal monitoring method for industrial water and electricity energy consumption, applied to electronic devices, characterized in that, The method includes: Periodically obtaining the energy consumption of industrial water and the energy consumption of industrial electricity to obtain a water energy sequence and an electric energy sequence; Based on the water energy sequence and the electric energy sequence, determining corresponding water energy characteristics and electric energy characteristics; According to the water energy characteristics and the electric energy characteristics, constructing a hydroelectricity correlation feature; When it is determined that the water energy sequence and the electric energy sequence are normal according to the hydroelectricity correlation feature, processing the water energy sequence and the electric energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electric energy; Based on the predicted water energy and the predicted electric energy, outputting an anomaly monitoring result of the water energy sequence and the electric energy sequence.

2. The industrial water and electricity energy consumption abnormal monitoring method according to claim 1, wherein The constructing the hydroelectricity correlation feature according to the water energy characteristics and the electric energy characteristics includes: Determining a hydroelectricity coupling coefficient of the water energy characteristics and the electric energy characteristics according to the power data in the electric energy characteristics, the flow data and the pressure difference data in the water energy characteristics; Determining an energy consumption matching degree of the water energy characteristics and the electric energy characteristics according to the first actual energy consumption and the first planned energy consumption in the water energy characteristics and the second actual energy consumption and the second planned energy consumption in the electric energy characteristics; According to the timestamp of the electric energy characteristics, determining the peak-time energy consumption and the valley-time energy consumption from the second actual energy consumption; Determining an energy consumption distribution coefficient of the electric energy characteristics according to the peak-time energy consumption and the valley-time energy consumption; Determining an indirect correlation coefficient of the water energy characteristics and the electric energy characteristics according to the energy consumption distribution coefficient and the energy consumption matching degree; Determining the hydroelectricity correlation feature according to the hydroelectricity coupling coefficient and the indirect correlation coefficient.

3. The industrial water and electricity energy consumption anomaly monitoring method according to claim 2, characterized in that, The determining the hydroelectricity coupling coefficient of the water energy characteristics and the electric energy characteristics includes: Determining the product of the flow data and the pressure difference data; Determining the hydroelectricity coupling coefficient according to the ratio of the power data to the product; wherein, the hydroelectricity coupling coefficient is used to indicate the correlation degree between the water energy characteristics and the electric energy characteristics.

4. The industrial water and electricity energy consumption anomaly monitoring method according to claim 2, characterized in that, The determining the indirect correlation coefficient of the water energy characteristics and the electric energy characteristics according to the energy consumption distribution coefficient and the energy consumption matching degree includes: Determining the product of the energy consumption distribution coefficient and the energy consumption matching degree as the indirect correlation coefficient of the water energy characteristics and the electric energy characteristics.

5. The industrial hydroelectric energy consumption anomaly monitoring method according to claim 2, wherein Determining that the water energy sequence and the electric energy sequence are abnormal according to the hydroelectricity correlation feature includes: Constructing nodes according to the water energy characteristics and the electric energy characteristics; Determining the edge weights between any two nodes according to the hydroelectricity coupling coefficient and the indirect correlation coefficient; Constructing hydroelectricity graph data according to the nodes and the edge weights; Based on a pre-trained graph neural network, determining an encoded vector of the hydroelectricity graph data; When the difference degree between the encoded vector and a pre-stored historical encoded vector is greater than or equal to a preset difference threshold, determining that the water energy sequence and the electric energy sequence are abnormal; When the difference degree between the encoded vector and a pre-stored historical encoded vector is less than the preset difference threshold, determining that the water energy sequence and the electric energy sequence are normal.

6. The industrial water and electricity energy consumption anomaly monitoring method according to claim 1, characterized in that The method further includes training the spatio-temporal prediction model, and the training the spatio-temporal prediction model includes: Construct a first time-domain encoding and a second time-domain encoding based on the pre-acquired historical water energy sequence, historical electric energy sequence, and corresponding timestamps. Construct a first spatial-domain encoding and a second spatial-domain encoding based on the historical water energy sequence, the historical electric energy sequence, and corresponding location information. Based on the first time-domain encoding and the first spatial-domain encoding, determine corresponding predicted time-domain encoding and predicted spatial-domain encoding according to a pre-constructed initial prediction model. Determine a first loss value of the initial prediction model according to the difference between the second time-domain encoding and the predicted time-domain encoding, and the difference between the second spatial-domain encoding and the predicted spatial-domain encoding. Update the initial prediction model according to the backpropagation algorithm until the first loss value meets a preset condition, and stop updating the initial prediction model to obtain a spatio-temporal prediction model trained to a convergent state.

7. The industrial water and electricity energy consumption anomaly monitoring method according to claim 6, characterized in that, The method for determining the first loss value of the initial prediction model satisfies the following first loss function: ； Among them, Loss represents the first loss value; The first weight parameter representing the difference between the second time-domain coding and the predicted time-domain coding; The second weight parameter representing the difference between the second spatial-domain coding and the predicted spatial-domain coding; The third weight parameter representing the similarity between the second time-domain coding and the second spatial-domain coding; i represents the index of the dimension in the second time-domain coding, n represents the number of dimensions in the second time-domain coding; j represents the index of the dimension in the second spatial-domain coding, m represents the number of dimensions in the second spatial-domain coding; Represents the value of the i-th dimension in the second time-domain coding; Represents the value of the i-th dimension in the predicted time-domain coding; Represents the value of the j-th dimension in the second spatial-domain coding; Represents the value of the j-th dimension in the predicted spatial-domain coding; G represents the similarity between the second time-domain coding and the second spatial-domain coding.

8. The industrial water and electricity energy consumption abnormal monitoring method according to claim 7, characterized in that The method further includes: When the number of times of determining predicted water energy and predicted electric energy based on the spatio-temporal prediction model reaches a preset number threshold, determine the water energy sequence and electric energy sequence corresponding to the predicted water energy and predicted electric energy output by the spatio-temporal prediction model last time. Update the first loss function based on the similarity between the water energy sequence and the electric energy sequence.

9. An industrial water and electricity energy consumption abnormal monitoring system, characterized in that, The system includes an electronic device and a sensor; The electronic device is used to periodically obtain industrial water energy consumption and industrial electricity energy consumption from the sensor to obtain a water energy sequence and an electric energy sequence; The electronic device is further used to determine corresponding water energy characteristics and electric energy characteristics based on the water energy sequence and the electric energy sequence; The electronic device is further used to construct a hydroelectricity correlation feature according to the water energy characteristics and the electric energy characteristics; The electronic device is further used to, when it is determined that there is no abnormality in the water energy sequence and the electric energy sequence according to the hydroelectricity correlation feature, process the water energy sequence and the electric energy sequence according to a pre-trained spatio-temporal prediction model to obtain predicted water energy and predicted electric energy; The electronic device is further used to output an abnormality monitoring result of the water energy sequence and the electric energy sequence based on the predicted water energy and the predicted electric energy.

10. The industrial water and electricity energy consumption anomaly monitoring system according to claim 9, characterized in that, The electronic device constructs the hydroelectricity correlation feature according to the water energy characteristics and the electric energy characteristics, including: Determine a hydroelectricity coupling coefficient of the water energy characteristics and the electric energy characteristics according to the power data in the electric energy characteristics, the flow data and pressure difference data in the water energy characteristics; Determine an energy consumption matching degree of the water energy characteristics and the electric energy characteristics according to the first actual energy consumption and the first planned energy consumption in the water energy characteristics, and the second actual energy consumption and the second planned energy consumption in the electric energy characteristics; Determine peak-time energy consumption and valley-time energy consumption from the second actual energy consumption according to the timestamp of the electric energy characteristics; Determine an energy consumption distribution coefficient of the electric energy characteristics according to the peak-time energy consumption and the valley-time energy consumption; Determine an indirect correlation coefficient of the water energy characteristics and the electric energy characteristics according to the energy consumption distribution coefficient and the energy consumption matching degree; Determine the hydroelectricity correlation feature according to the hydroelectricity coupling coefficient and the indirect correlation coefficient.

Images