Industrial enterprise energy efficiency diagnosis method based on big data analysis technology

Abstract

The invention discloses an industrial enterprise energy efficiency diagnosis method based on a big data analysis technology. According to the method, a fusion data set is constructed by integrating multi-source heterogeneous data of energy, production, environment, management and the like; further defining and calculating energy efficiency key performance indicators; a dynamic energy consumption reference model is established through machine learning, and real-time energy efficiency abnormity diagnosis is achieved; performing root cause analysis through a data mining algorithm, and positioning a problem source; and the energy-saving potential is quantitatively evaluated, and finally the diagnosis result and the optimization suggestion are visually output. According to the invention, closed-loop management from monitoring to diagnosis optimization is realized, the defects of strong subjectivity, data islanding and lack of intelligent analysis in the prior art are overcome, and the method has the advantages of being comprehensive, accurate, real-time and strong in operability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial energy saving, in particular to an industrial enterprise energy efficiency diagnosis method based on big data analysis technology. BACKGROUND

[0002] The energy consumption management of industrial enterprises is increasingly becoming the key to reduce costs and achieve green development. Energy efficiency diagnosis is the first step to find energy waste and tap energy saving potential.

[0003] Currently, the energy efficiency diagnosis of industrial enterprises mainly relies on two ways: one is the traditional manual diagnosis method, which is to send a team of energy audit experts to the enterprise, and to conduct preliminary analysis through on-site investigation, inquiry, and review of historical energy bills and reports. This method highly depends on the personal experience of experts, has long diagnosis cycle and high cost, and is difficult to find deep and dynamic energy efficiency problems, and has strong subjectivity, low efficiency, lack of real-time, and inability to handle massive data. The second is a simple monitoring method based on SCADA / DCS system, which can collect and display the energy consumption of key equipment, but mostly only has data dashboard and threshold alarm functions. Its defects are that it only monitors but does not controls, and only reports but does not diagnoses, and cannot automatically analyze the reasons for high energy consumption; at the same time, energy consumption data, production data, and environmental data are scattered in different systems, forming data islands, and lacking effective correlation and fusion analysis; in addition, the existing system lacks intelligent analysis capability and cannot use historical data to build a prediction model for energy efficiency benchmark comparison, root cause analysis, and energy saving potential prediction.

[0004] Therefore, the existing technology urgently needs a method that can automatically, intelligently, comprehensively, and deeply diagnose and analyze the energy efficiency of industrial enterprises. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art and provide an industrial enterprise energy efficiency diagnosis method based on big data analysis technology. This method can integrate multi-source heterogeneous data, establish a data-driven model, and realize automatic, real-time, and accurate diagnosis and optimization potential evaluation of energy efficiency.

[0006] In order to solve the above problems, the present application discloses an industrial enterprise energy efficiency diagnosis method based on big data analysis technology, comprising the following steps: S1, multi-source heterogeneous data collection and fusion: collecting energy data, production and operation data, environmental data, and enterprise information data from multiple information systems of the enterprise, and performing cleaning, timestamp alignment, and labeling processing on the collected raw data to form a fusion data set; S2, energy efficiency key performance indicator (KPI) system construction and calculation: defining and calculating a series of energy efficiency KPIs based on the fusion data set; S3, Energy efficiency modeling and analysis driven by big data: based on the fusion data set and energy efficiency KPI, the following sub-steps are performed: S3.1, using machine learning algorithm, based on historical normal working condition data, establishing expected energy consumption benchmark model under different production conditions and environmental conditions; S3.2, input real-time data into the benchmark model, calculate the expected energy consumption value, and compare with the measured energy consumption value, if the deviation exceeds the preset threshold, trigger energy efficiency abnormal alarm; S3.3, for the triggered energy efficiency anomaly, the data mining algorithm is used to analyze the production or environmental parameters with the highest correlation, and the root cause analysis is carried out; S3.4, based on model and data analysis, quantitatively evaluate the energy saving and cost saving that can be achieved after adjusting the abnormal parameters to the optimal interval; S4, visualization of diagnosis results and decision support: the energy efficiency KPI, energy efficiency abnormal alarm, root cause analysis results and energy saving potential evaluation results are visualized and displayed, and diagnosis report and optimization suggestion are generated.

[0007] Preferably, in the step of multi-source heterogeneous data collection and fusion, the energy data includes the consumption data of electricity, gas, water and steam; the production and operation data includes the device start-stop state, running speed, production line yield, temperature, pressure, flow process parameters; the environmental data includes the environmental temperature and humidity; the enterprise information data includes production plan, shift information and product specifications.

[0008] Preferably, the energy efficiency key performance indicator KPI includes unit product energy consumption, equipment overall efficiency OEE, load rate and system operation efficiency.

[0009] Preferably, the machine learning algorithm used to establish the expected energy consumption benchmark model includes regression model or time series analysis algorithm.

[0010] Preferably, the data mining algorithm used for root cause analysis includes correlation analysis, clustering analysis or decision tree algorithm.

[0011] Preferably, in the step of visualization of diagnosis results and decision support, the visualization is realized through mobile terminal, and the display content includes enterprise energy flow diagram, real-time energy efficiency KPI board, energy efficiency abnormal alarm list and root cause analysis report, energy saving potential evaluation report and optimization suggestion.

[0012] Compared with the prior art, the beneficial effects of the present application are: 1. comprehensiveness and systematicness: breaking the data island, deeply fusing energy, production and environmental data, and diagnosing energy efficiency from the whole enterprise and system level.

[0013] 2. Intelligence and Precision: By utilizing big data analysis and machine learning models to replace traditional human experience, it can discover complex relationships and deep-seated problems from massive amounts of data, resulting in more objective and accurate diagnostic results.

[0014] 3. Foresight and predictive: Energy efficiency benchmark models can be used for post-event diagnosis, predict future energy consumption trends, evaluate energy efficiency performance under different decision-making schemes, and achieve pre-event optimization.

[0015] 4. High efficiency and real-time performance: The diagnostic process is automated, enabling near real-time energy efficiency monitoring and anomaly alarms, which greatly shortens the diagnostic cycle.

[0016] 5. Highly operable: It not only identifies problems, but also pinpoints the root causes through root cause analysis, and clarifies energy-saving potential and economic benefits through quantitative assessment, providing specific and actionable optimization suggestions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the energy efficiency diagnosis method of the present invention. Detailed Implementation

[0019] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0021] The present invention will now be described in further detail with reference to the accompanying drawings.

[0022] like Figure 1As shown, the implementation process of this invention begins with the collection and fusion of multi-source heterogeneous data, followed by the calculation of energy efficiency KPIs, and then enters the core modeling and analysis stage (including benchmark modeling, anomaly diagnosis, root cause analysis and potential assessment). Finally, the results are visualized and used for decision support, forming a complete closed loop.

[0023] Taking a certain injection molding factory as an example, the specific implementation of the present invention will be described.

[0024] (a) Data collection: (1) Collect electricity consumption data (energy data) of the entire plant and each injection molding machine from smart meters.

[0025] (2) Collect process parameters (production data) such as the operating status, hydraulic pressure, and barrel temperature of each injection molding machine from the SCADA system.

[0026] (3) Collect production product models and output data (production data) for each shift and each piece of equipment from the MES system.

[0027] (4) Collect workshop environmental data from temperature and humidity sensors.

[0028] All data is uploaded to the cloud platform or local server's data lake through the data acquisition gateway, and then synchronized and cleaned.

[0029] (II) KPI Calculation: The system automatically calculates the unit product power consumption (kWh / kg) for each injection molding machine when producing a specific product.

[0030] Calculate the overall energy intensity of the entire plant.

[0031] (III) Energy Efficiency Modeling and Analysis: (1) Establishing a benchmark model: The system uses data from the past 3 months of normal production of product A to train a multiple linear regression model. The input variables of the model include: ambient temperature, set output, barrel temperature setpoint, etc.; the output variable is the expected energy consumption.

[0032] (2) Abnormal diagnosis: On a certain day, the system found that when injection molding machine No. 5 was producing product A, the actual energy consumption was 15% higher than the model prediction value, which lasted for more than 1 hour, and then triggered a yellow alarm.

[0033] (3) Root cause analysis: The system automatically ran the root cause analysis algorithm and found that the abnormal energy consumption of the machine was highly negatively correlated with the parameter of "cooling water valve opening" (the energy consumption increased when the valve opening decreased). Further inspection revealed that the valve failed to open completely due to a malfunction, resulting in a decrease in cooling efficiency. In order to maintain the barrel temperature, the heater continued to work at high power, causing an increase in power consumption.

[0034] (4) Potential assessment: The system assessment report indicates that after the valve is repaired, the energy consumption of the machine for producing product A can be restored to normal levels, and it is estimated that electricity costs can be saved by about 2,000 yuan per month.

[0035] (iv) Visualization and decision support: The aforementioned alarm information, root cause analysis reports, and energy-saving potential assessments are all displayed on the large screen in the factory's energy management center.

[0036] The system simultaneously pushed a maintenance work order to the equipment maintenance engineer's mobile app: "Suspected malfunction of the cooling water valve on injection molding machine No. 5; immediate inspection recommended." After the engineer confirmed the issue on-site and replaced the valve, energy efficiency returned to normal, the system deactivated the alarm, and recorded the incident and energy-saving results.

[0037] Through the above implementation methods, the present invention achieves automated, intelligent, and refined diagnosis and management of energy efficiency in industrial enterprises.

[0038] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for diagnosing energy efficiency in industrial enterprises based on big data analytics, characterized in that, Includes the following steps: S1. Multi-source heterogeneous data acquisition and fusion: Collect energy data, production and operation data, environmental data and enterprise information data from multiple information systems of the enterprise, and clean, timestamp-align and label the collected raw data to form a fused dataset; S2. Construction and calculation of energy efficiency key performance indicator (KPI) system: Based on the fused dataset, define and calculate a series of energy efficiency KPIs; S3. Big Data-Driven Energy Efficiency Modeling and Analysis: Based on the fused dataset and energy efficiency KPIs, perform the following sub-steps: S3.1 Utilize machine learning algorithms to establish a benchmark model of expected energy consumption under different production conditions and environmental conditions based on historical normal operating data; S3.2 Input the real-time data into the benchmark model, calculate the expected energy consumption value, and compare it with the measured energy consumption value. If the deviation exceeds the preset threshold, trigger an energy efficiency anomaly alarm. S3.3 For triggered energy efficiency anomalies, use data mining algorithms to analyze the production or environmental parameters most correlated with them and conduct root cause analysis. S3.

4. Based on model and data analysis, quantitatively evaluate the energy saving and cost savings that can be achieved after adjusting abnormal parameters to the optimal range; S4. Visualization of Diagnostic Results and Decision Support: Visualize energy efficiency KPIs, energy efficiency anomaly alarms, root cause analysis results, and energy-saving potential assessment results, and generate diagnostic reports and optimization suggestions.

2. The method for diagnosing energy efficiency in industrial enterprises based on big data analytics technology according to claim 1, characterized in that: In the multi-source heterogeneous data acquisition and fusion process, energy data includes consumption data of electricity, gas, water, and steam; production and operation data includes equipment start-up and shutdown status, operating rate, production line output, temperature, pressure, flow rate, and process parameters; environmental data includes ambient temperature and humidity; and enterprise information data includes production plans, shift information, and product specifications.

3. The method for diagnosing energy efficiency in industrial enterprises based on big data analytics technology according to claim 1, characterized in that: Energy efficiency key performance indicators (KPIs) include energy consumption per unit product, overall equipment efficiency (OEE), load factor, and system operating efficiency.

4. A method for diagnosing energy efficiency in industrial enterprises based on big data analytics, as claimed in claim 1, is characterized in that: The machine learning algorithms used to establish the expected energy consumption benchmark model include regression models or time series analysis algorithms.

5. A method for diagnosing energy efficiency in industrial enterprises based on big data analytics, as claimed in claim 1, characterized in that: Data mining algorithms used for root cause analysis include correlation analysis, cluster analysis, or decision tree algorithms.

6. A method for diagnosing energy efficiency in industrial enterprises based on big data analytics, as claimed in claim 1, characterized in that: In the diagnostic results visualization and decision support step, the visualization is achieved through mobile devices. The displayed content includes the enterprise's energy flow diagram, real-time energy efficiency KPI dashboard, energy efficiency anomaly alarm list and root cause analysis report, energy saving potential assessment report and optimization suggestions.

Images