Operation equipment energy consumption optimization management method based on big data analysis

Abstract

The invention provides an operation equipment energy consumption optimization management method based on big data analysis, and the method comprises the steps: forming sequence energy consumption data through continuously obtaining the energy consumption data of each operation equipment group for producing a single product; constructing an energy consumption matrix by taking the sequence energy consumption data of each operation equipment group as matrix elements; obtaining the principal component of the energy consumption matrix by adopting a principal component analysis method, and obtaining the contribution rate of each operation equipment group to the principal component; determining the operation equipment group with the contribution rate higher than a preset value as an abnormal operation equipment group; inputting the sequence energy consumption data of the abnormal operation equipment group into a preset LSTM model, and outputting an energy consumption abnormal feature; and managing the start and stop of each operation equipment group according to the energy consumption abnormal characteristics. According to the method, the problem of low energy consumption management efficiency caused by difficulty in scheduling and energy consumption abnormity checking according to the energy consumption difference of the running equipment is solved, the equipment abnormity is found through multi-dimensional data comparison and timely maintenance is performed, and the production cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of operating equipment management, and in particular to a method for optimizing the energy consumption of operating equipment based on big data analysis. Background Art

[0002] In the industrial production sector, equipment energy management is a crucial component of enterprise operations. Currently, traditional approaches to equipment energy management suffer from numerous shortcomings. For one thing, the lack of in-depth analysis of equipment energy consumption data makes it difficult to accurately identify potential issues during equipment operation. For example, during production, energy consumption differences between the same equipment under different production conditions, or between different equipment under the same conditions, cannot be captured promptly and accurately. This results in potential equipment failures being delayed, impacting the overall operational efficiency and service life of the equipment.

[0003] On the other hand, equipment operation scheduling and energy utilization are not fully optimized by integrating external factors. Under the current model, there is a lack of systematic planning for the utilization of peak and off-peak electricity price periods. Equipment operating hours are not properly adjusted based on the actual characteristics of the equipment and production rhythm, making it difficult to improve energy efficiency. In terms of enterprise circuit capacity management, the control method is rigid, and it is impossible to flexibly adjust equipment operating status during peak electricity consumption periods based on real-time production needs and equipment energy consumption. This easily leads to waste of electricity resources and even affects the stable operation of equipment due to excessive instantaneous load, increasing the risk of equipment failure. Summary of the Invention

[0004] An embodiment of the present invention provides a method for optimizing the energy consumption management of operating equipment based on big data analysis, aiming to solve the problem that the existing technology is difficult to schedule and troubleshoot energy consumption anomalies based on the energy consumption differences of operating equipment, resulting in low energy consumption management efficiency.

[0005] To achieve the above objectives, the present invention provides a method for optimizing the energy consumption of operating equipment based on big data analysis, comprising the following steps: By continuously acquiring the energy consumption data of each operating equipment group in producing a single product, serial energy consumption data is formed; The energy consumption matrix is ​​constructed by taking the serial energy consumption data of each operating equipment group as matrix elements; Using a principal component analysis method to obtain the principal components of the energy consumption matrix, and obtaining the contribution rate of each operating equipment group to the principal components; Determine the operating equipment group whose contribution rate is higher than a preset value as an abnormal operating equipment group; Inputting the sequence energy consumption data of the abnormally operating equipment group into a preset LSTM model and outputting energy consumption abnormality features; The start and stop of each operating equipment group is managed according to the abnormal energy consumption characteristics.

[0006] Furthermore, the energy consumption data of each operating equipment group producing a single product is continuously obtained, specifically: electricity consumption data is collected through a smart meter installed at the power supply end of the equipment, gas consumption data of gas-consuming equipment is collected through a gas flow meter, and liquid consumption data of oil-consuming equipment is collected through an oil consumption sensor.

[0007] Furthermore, it also includes: collecting equipment operating status parameters through temperature sensors, pressure sensors and flow sensors, and correlating and analyzing the operating status parameters with energy consumption data.

[0008] Furthermore, in the energy consumption matrix, the row elements of the matrix correspond to different operating equipment groups, and the column elements of the matrix correspond to different production sequences or time nodes of the products.

[0009] Furthermore, the preset value is determined based on the statistical distribution of historical energy consumption data, specifically by calculating the mean and standard deviation of the contribution rate of each operating equipment group in the historical data, and taking the mean plus K times the standard deviation as the preset value.

[0010] Furthermore, the preset LSTM model is trained through the following steps: obtaining historical abnormal energy consumption data as training samples, and dividing the training samples into a training set and a test set; using the sequence energy consumption data in the training set as the input of the LSTM model, and the corresponding preset energy consumption abnormality features as the output labels, and using the training set to train the LSTM model; using the test set to verify and adjust the parameters of the trained LSTM model.

[0011] Furthermore, the start and stop of each operating equipment group also includes: obtaining the peak and valley electricity price period data of the power grid, and associating the peak and valley electricity price period data with the serial energy consumption data of each operating equipment group; when it is detected that it is in the peak and valley electricity price period, adjusting the start and stop time of the operating equipment group according to the preset delayed start strategy or early shutdown strategy.

[0012] Furthermore, the start and stop of each operating equipment group also includes: obtaining the grid capacity threshold data applied for by the enterprise, and monitoring the total power consumption capacity of each operating equipment group in real time; when the total power consumption capacity is monitored to exceed the preset capacity threshold, the abnormal operating equipment group is shut down first.

[0013] The above technical solution has the following technical effects: The energy consumption data of each operating equipment group producing a single product is continuously obtained to form serial energy consumption data; the serial energy consumption data of each operating equipment group is used as a matrix element to construct an energy consumption matrix; the principal component of the energy consumption matrix is ​​obtained using the principal component analysis method, and the contribution rate of each operating equipment group to the principal component is obtained; the operating equipment group with a contribution rate higher than a preset value is determined to be an abnormal operating equipment group; the serial energy consumption data of the abnormal operating equipment group is input into a preset LSTM model to output energy consumption anomaly characteristics; and the start and stop of each operating equipment group is managed based on the energy consumption anomaly characteristics. This invention solves the problem that the existing technology has difficulty in scheduling and troubleshooting energy consumption anomalies based on energy consumption differences of operating equipment, resulting in low energy consumption management efficiency.

[0014] In a further embodiment, by associating the peak and valley electricity price period data of the power grid with the equipment energy consumption data, and adjusting the equipment start and stop time, the equipment can be maximized to operate during the valley electricity price period, fully utilizing the low-price electricity period, effectively reducing the enterprise's electricity costs and improving economic benefits.

[0015] In a further embodiment, by real-time monitoring of the enterprise's electricity consumption capacity, when the total electricity consumption capacity exceeds the application threshold, combined with the abnormal energy consumption characteristics of the equipment and the criticality level of the production link, the equipment with abnormal energy consumption and non-critical links is shut down first to avoid additional costs caused by excessive capacity. Under the premise of ensuring production continuity, refined management of electricity consumption and cost optimization are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of a method for optimizing energy consumption management of operating equipment based on big data analysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate various embodiments, the present invention is provided with accompanying drawings. These drawings form part of the present disclosure and are primarily used to illustrate the embodiments and, in conjunction with the relevant description in the specification, to explain the operating principles of the embodiments. By referring to these drawings, one of ordinary skill in the art will understand other possible embodiments and the advantages of the present invention. The components in the figures are not drawn to scale, and similar reference numerals are generally used to represent similar components.

[0018] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0019] Figure 1 FIG. 1 is a flow chart of a method for optimizing energy consumption management of operating equipment based on big data analysis according to an embodiment of the present invention. Figure 1 As shown, the method of this embodiment includes the following steps: By continuously acquiring the energy consumption data of each operating equipment group in producing a single product, serial energy consumption data is formed; In one specific implementation, in order to obtain the energy consumption data of a single product produced by a group of operating equipment to form a series of energy consumption data, it is necessary to use professional equipment to achieve accurate collection. In terms of power consumption data collection, a smart meter is installed at the power supply end of the equipment. It can use the internal metering unit to monitor key parameters such as current, voltage, power factor, and power consumption during equipment operation in real time. For example, in an industrial production scenario, connecting a smart meter to the power line of a large machine tool can accurately obtain the power consumption of each component when the machine tool is processing it. This data is transmitted to the data processing unit in real time.

[0020] For gas-consuming equipment, gas flow meters are used to collect gas consumption data. Gas turbine flow meters and intelligent vortex gas flow meters are widely used. They measure gas flow based on different principles. Intelligent vortex gas flow meters, for example, are based on the principle of vortex precession and have built-in intelligent chips. They offer functions such as automatic diagnosis and temperature and pressure compensation. They can automatically collect and store data without an external power supply, and display parameters such as flow rate and cumulative volume in real time on an LCD screen. They are suitable for measuring various gases, including natural gas and city gas, and can accurately measure the amount of gas consumed by gas-consuming equipment when producing a single product.

[0021] For fuel-consuming equipment, fuel consumption sensors are used to collect liquid consumption data. For example, an automotive fuel consumption meter uses a precision flow sensor to convert incoming fuel into an electrical signal. This information is then processed and displayed by a central processing unit (CPU). This data allows for real-time monitoring and display of both instantaneous and cumulative fuel consumption, as well as accurate fuel consumption data for individual products produced by fuel-consuming equipment. This energy consumption data, collected from various devices and arranged in chronological order, forms sequential energy consumption data, laying a solid foundation for in-depth analysis and management of equipment energy consumption.

[0022] Taking mobile phone case production as an example, obtaining energy consumption data for groups of operating equipment and constructing sequential energy consumption data requires the use of a variety of sensors for precise collection. During the injection molding process, a smart meter installed at the equipment power supply monitors the current, voltage, and power consumption of the injection molding machine in real time as it produces each mobile phone case, accurately recording the power consumption of each injection operation. For surface treatment processes, a gas flow meter monitors the gas consumption of sandblasting and spraying equipment, such as the real-time use of compressed air during cleaning or coating of mobile phone cases. For hydraulically driven mobile phone case stamping equipment, an oil consumption sensor measures hydraulic oil consumption and records the fuel usage for producing a single mobile phone case. These measurement data, from different sensors and targeting different types of energy consumption, are arranged in chronological order to form a comprehensive sequential energy consumption data set that reflects the mobile phone case production process.

[0023] In one specific implementation, temperature sensors, pressure sensors, and flow sensors are used to collect equipment operating status parameters, which are then correlated with energy consumption data for analysis. Temperature sensors monitor real-time temperature changes in key equipment components, such as motors and bearings. Excessively high temperatures are often accompanied by abnormally increased energy consumption, potentially indicating excessive equipment load or component wear. Pressure sensors monitor the internal pressure of equipment such as pipelines and pressure vessels. Abnormal pressure fluctuations not only affect production efficiency but can also lead to increased energy consumption. Flow sensors accurately measure the flow rate of liquids or gases. For conveying equipment such as pumps and fans, flow parameters are directly correlated with energy consumption. By correlating these operating status parameters with energy consumption data, it is possible to further identify potential causes of abnormal energy consumption. For example, if a sudden increase in energy consumption is detected for a piece of equipment, combined with component overheating data displayed by the temperature sensor, possible heat dissipation failures or mechanical friction issues can be quickly identified. Alternatively, if the pressure value reported by the pressure sensor exceeds the normal range and energy consumption data also increases significantly, it can be determined that the equipment is operating inefficiently. This cross-validation of multi-dimensional data not only improves the accuracy of anomaly identification, but also provides a scientific basis for equipment optimization, fault warning, and maintenance strategy formulation.

[0024] The energy consumption matrix is ​​constructed by taking the serial energy consumption data of each operating equipment group as matrix elements; In a specific implementation, the row elements of the matrix correspond to different groups of operating equipment, and the column elements of the matrix correspond to different production sequences or time nodes of the products. When constructing the serial energy consumption data of each operating equipment group into a matrix, the energy consumption data of each product is used as the matrix element. The core is to structure and integrate discrete energy consumption information. Specifically, the rows of the matrix correspond to different groups of operating equipment, reflecting the energy consumption differences between the equipment; the columns correspond to different product production sequences or time nodes, reflecting the dynamic changes in energy consumption. This construction method allows each matrix element to accurately correspond to the energy consumption data of a single product produced by a certain equipment group, and fully retains the dual characteristics of equipment energy consumption in the time dimension and the equipment dimension. Based on this matrix, the energy consumption patterns can be systematically mined through methods such as principal component analysis, and equipment groups with abnormal energy consumption can be identified, providing a scientific data basis for energy consumption optimization, and realizing an effective transition from individual product energy consumption data to global equipment energy consumption management.

[0025] The principal component analysis method is used to obtain the principal components of the energy consumption matrix and the contribution rate of each operating equipment group to the principal components; Determine the operating equipment group whose contribution rate is higher than the preset value as the abnormal operating equipment group; In a specific implementation, the preset value is determined based on the statistical distribution of historical energy consumption data, specifically: calculating the mean and standard deviation of the contribution rate of each operating equipment group in the historical data, and taking the mean plus K times the standard deviation as the preset value.

[0026] In energy consumption matrix analysis, principal component analysis (PCA) uses dimensionality reduction to extract key features (principal components) that explain energy consumption fluctuations. The contribution rate of each equipment group to the principal components is then calculated. In specific implementation, the energy consumption matrix is ​​first normalized to eliminate dimensionality effects. The eigenvalues ​​and eigenvectors of the covariance matrix are then calculated, and the principal components are determined based on the eigenvalue ratios. The contribution rate reflects the degree of influence of the equipment group's energy consumption fluctuations on the principal components. For example, the higher the contribution rate of a particular equipment group, the more dominant its energy consumption changes are in relation to the overall energy consumption trend.

[0027] The preset value is dynamically determined based on the statistical distribution of historical data: the mean and standard deviation of each device group's contribution rate are calculated, and the threshold is "mean + K times the standard deviation." In one specific implementation, K is typically set between 1 and 3, adjusted based on the degree of data dispersion. Based on statistical principles, this method identifies contribution rates that fall outside the normal fluctuation range as abnormal. For example, if a device group's contribution rate exceeds the mean + 2 times the standard deviation, its energy consumption fluctuations significantly deviate from historical norms, thus identifying it as an abnormally operating device group. This dynamic threshold setting avoids subjective experience bias, ensuring that anomaly identification more closely matches the actual operating patterns of the devices, and providing accurate screening for abnormal devices for subsequent LSTM model analysis.

[0028] Input the sequence energy consumption data of the abnormally operating equipment group into the preset LSTM model and output the energy consumption abnormality characteristics; In one specific implementation, a preset LSTM model is trained through the following steps: obtaining historical abnormal energy consumption data as training samples and dividing the training samples into a training set and a test set; using the sequential energy consumption data in the training set as input to the LSTM model, and using the corresponding preset energy consumption anomaly features as output labels, the LSTM model is trained using the training set; and the trained LSTM model is validated and parameterized using the test set. In one specific implementation, the energy consumption anomaly features include sudden increases in energy consumption and / or abnormal energy consumption fluctuations.

[0029] The start and stop of each operating equipment group is managed according to the abnormal energy consumption characteristics.

[0030] In a specific implementation, the start and stop of each operating equipment group also includes: obtaining the peak and valley electricity price period data of the power grid, and associating the peak and valley electricity price period data with the serial energy consumption data of each operating equipment group; when it is detected that it is in the peak and valley electricity price period, adjusting the start and stop time of the operating equipment group according to the preset delayed start strategy or early shutdown strategy.

[0031] In equipment start-up and shutdown management, the core logic of optimizing energy consumption costs by combining the peak-valley electricity price mechanism is to achieve peak-shifting management of electricity consumption periods through data association and strategy adjustment. Specifically, first obtain the time period division data of the power grid's peak-valley electricity price, such as the valley floor price is 24:00-7:00 and the valley peak price is 7:00-24:00, and then associate and match it with the serial energy consumption data of each equipment group in the time dimension. When the system detects that the equipment is operating during the valley-peak electricity price period, that is, the high electricity price range, it triggers the preset start-up and shutdown adjustment strategy: Delayed start strategy: This strategy shifts the equipment's original off-peak start time to the beginning of the off-peak electricity price period. For example, a device originally scheduled to start at 8:00 AM is adjusted to 6:00 AM. The device is then preheated or in standby mode during the off-peak period, allowing it to resume production when normal operation begins. This shifts the high-energy preheating phase to the lower-price electricity period.

[0032] Early shutdown strategy: Shut down equipment before the peak electricity price period begins to avoid operating during high electricity price periods. For example, if the equipment is originally scheduled to shut down at 24:00, if the peak electricity price period begins at 24:00, shut it down at 23:50, shifting the final stage of operation energy consumption to the off-peak period.

[0033] This approach allows equipment to operate at peak energy consumption levels during off-peak electricity price periods, reducing unit energy costs. The implementation of this strategy relies on the precise matching of sequential energy consumption data with electricity price periods, as well as pre-setting parameters such as equipment warm-up time and standby energy consumption, ultimately achieving a balance between economic efficiency and production efficiency.

[0034] In a specific implementation, the start and stop of each operating equipment group also includes: obtaining the grid capacity threshold data applied for by the enterprise, and monitoring the total power consumption capacity of each operating equipment group in real time; when the total power consumption capacity is monitored to exceed the preset capacity threshold, the abnormal operating equipment group is shut down first.

[0035] In equipment start-up and shutdown management, the core logic of energy consumption optimization combined with grid capacity thresholds is to avoid excessive costs caused by exceeding power capacity limits through real-time monitoring and dynamic shutdown strategies. In specific implementation, the grid capacity threshold applied by the enterprise to the power bureau, such as 1000kVA, is first obtained. The power consumption data of each equipment group is collected in real time through the power monitoring system to calculate the total power consumption capacity. When the total capacity exceeds the preset threshold, such as exceeding 1000kVA for 5 consecutive minutes, the system triggers the shutdown mechanism: Prioritize shutting down abnormally operating equipment groups, specifically those with abnormal energy consumption identified through principal component analysis and LSTM models. These devices often experience significant energy consumption fluctuations or malfunctions. Shutting down these devices can quickly reduce overall power consumption and minimize energy waste caused by abnormal energy consumption. For example, if a production line's total capacity exceeds its limit, one injection molding machine, identified as a high-contributing abnormal device, is found to be consuming 15% more energy than normal. Shutting down this device effectively reduces capacity load while avoiding the additional power consumption caused by abnormal operation.

[0036] The advantage of this strategy lies in combining the characteristics of abnormal equipment energy consumption with grid capacity management requirements, addressing capacity overloads while simultaneously optimizing equipment energy efficiency. After shutdown, total capacity is continuously monitored. When load drops below a threshold of 80%, equipment is restarted in order of production priority, ensuring that electricity costs are reduced while minimizing the impact on production continuity.

[0037] Although the present invention has been particularly shown and described in conjunction with preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made to the present invention without departing from the spirit and scope of the invention as defined in the appended claims, and all such changes are within the scope of protection of the present invention.

Claims

1. A method for optimizing the energy consumption of operating equipment based on big data analysis, characterized by: The following steps are involved: By continuously acquiring the energy consumption data of each operating equipment group in producing a single product, serial energy consumption data is formed; The energy consumption matrix is ​​constructed by taking the serial energy consumption data of each operating equipment group as matrix elements; Using a principal component analysis method to obtain the principal components of the energy consumption matrix, and obtaining the contribution rate of each operating equipment group to the principal components; Determine the operating equipment group whose contribution rate is higher than a preset value as an abnormal operating equipment group; Inputting the sequence energy consumption data of the abnormally operating equipment group into a preset LSTM model and outputting energy consumption abnormality features; The start and stop of each operating equipment group is managed according to the abnormal energy consumption characteristics.

2. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The continuous acquisition of energy consumption data of each operating equipment group producing a single product is specifically: collecting electricity consumption data through a smart meter installed at the power supply end of the equipment, collecting gas consumption data of gas-consuming equipment through a gas flow meter, and collecting liquid consumption data of oil-consuming equipment through an oil consumption sensor.

3. The method for optimizing energy consumption of operating equipment based on big data analysis according to claim 2, characterized in that: The equipment operating status parameters are collected through temperature sensors, pressure sensors and flow sensors, and the operating status parameters are correlated with energy consumption data for analysis.

4. The method for optimizing energy consumption of operating equipment based on big data analysis according to claim 1 is characterized in that: In the energy consumption matrix, the row elements of the matrix correspond to different operating equipment groups, and the column elements of the matrix correspond to different production sequences or time nodes of the products.

5. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The preset value is determined based on the statistical distribution of historical energy consumption data, specifically by calculating the mean and standard deviation of the contribution rate of each operating equipment group in the historical data, and taking the mean plus K times the standard deviation as the preset value.

6. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The preset LSTM model is trained by the following steps: obtaining historical abnormal energy consumption data as training samples, dividing the training samples into a training set and a test set; using the sequence energy consumption data in the training set as the input of the LSTM model, and the corresponding preset energy consumption abnormality features as the output labels, and using the training set to train the LSTM model; using the test set to verify and adjust the parameters of the trained LSTM model.

7. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The start and stop of each operating equipment group also includes: obtaining the peak and valley electricity price period data of the power grid, and associating the peak and valley electricity price period data with the serial energy consumption data of each operating equipment group; when it is detected that it is in the peak and valley electricity price period, adjusting the start and stop time of the operating equipment group according to the preset delayed start strategy or early shutdown strategy.

8. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The start and stop of each operating equipment group also includes: obtaining the grid capacity threshold data applied for by the enterprise, and monitoring the total power consumption capacity of each operating equipment group in real time; when the total power consumption capacity is monitored to exceed the preset capacity threshold, the abnormal operating equipment group is shut down first.

9. The method for optimizing energy consumption management of operating equipment based on big data analysis according to claim 1, characterized in that: The abnormal energy consumption characteristics include a sudden increase in energy consumption and / or abnormal energy consumption fluctuation amplitude.

Images