Internet-based enterprise energy use behavior intelligent analysis method and system

Abstract

The invention discloses an internet-based enterprise energy use behavior intelligent analysis method and system, and relates to the technical field of energy management. Energy consumption data is collected, and an energy data set S is constructed, so that a power consumption fluctuation index DL, a water resource consumption index WR and an equipment energy efficiency index NX are obtained; and performing comprehensive analysis according to the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX to obtain a comprehensive energy consumption score PF, evaluating the energy consumption state of the enterprise in combination with a preset energy consumption score threshold PFYZ, if the energy consumption state is evaluated as an abnormal state, triggering an early warning mechanism, and if the energy consumption state is evaluated as a normal state, predicting future energy consumption E, according to the method, energy optimization scheduling is carried out, energy consumption data after energy optimization scheduling is recorded, the energy optimization scheduling effect is evaluated, the energy utilization efficiency of an enterprise is improved, the operation cost of the enterprise is reduced, and scientific decision support is provided for energy conservation, consumption reduction and sustainable development of the enterprise.

Description

Technical Field

[0001] The present invention relates to the field of energy management technology, and in particular to an Internet-based intelligent analysis method and system for enterprise energy usage behavior. Background Art

[0002] As global energy issues become increasingly severe, corporate energy management has become a key link in improving operational efficiency and reducing costs. With the development of information technology, especially the popularization of Internet technology and big data analysis, corporate energy management has gradually developed in the direction of intelligence and automation. It collects energy consumption data of various equipment within the enterprise through the Internet, and uses intelligent analysis systems to comprehensively monitor and optimize energy consumption. It conducts data analysis on the energy use behavior of various equipment in the daily operation of the enterprise, predicts energy consumption, and optimizes energy use through intelligent scheduling, thereby achieving energy conservation and emission reduction, cost control and environmental protection.

[0003] In the Chinese invention patent with application publication number N112488558A, an energy consumption monitoring and analysis system based on the industrial Internet is disclosed. The system sets up a regional energy consumption data collection system and an enterprise energy consumption monitoring platform, uses the regional energy consumption data collection system to collect production data and energy consumption data of production equipment, unifies and standardizes the collected information and transmits the data to the enterprise energy consumption monitoring platform; and the enterprise energy consumption monitoring platform includes an online monitoring and intelligent management system for energy resource consumption of the management unit, which is used to store energy consumption data and compare it with the energy consumption quota value, energy consumption index value and energy consumption benchmark value of industrial enterprises in the region. It aims to solve the problems of high construction cost, weak system flexibility and poor sharing of power energy data processing methods in energy management in the existing technology, which are relatively common, and energy data cannot be sent to the person in charge of energy management in a timely and intuitive manner, especially in the energy consumption monitoring of enterprise production, and it is difficult to achieve a timely response.

[0004] The above system can solve the problems existing in the existing technology of high construction cost, weak system flexibility and poor sharing of power energy data processing methods in energy management. Energy data cannot be sent to the person in charge of energy management in a timely and intuitive manner, especially in the energy consumption monitoring of enterprise production, and it is difficult to respond in a timely manner.

[0005] However, this system lacks data processing and collection, has a single data dimension, and fails to conduct multi-dimensional correlation analysis and forecast future energy consumption. When faced with abnormal situations, the early warning mechanism is inflexible and it is difficult to provide detailed abnormal information and a graded early warning mechanism.

[0006] To this end, the present invention provides an Internet-based intelligent analysis method and system for enterprise energy usage behavior. Summary of the invention

[0007] In view of the deficiencies in the prior art, the present invention provides an Internet-based method and system for intelligent analysis of enterprise energy usage behavior, which solves the problems in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: an Internet-based enterprise energy usage behavior intelligent analysis system, including a data collection and preprocessing module, a data analysis module, an early warning module, an energy consumption prediction module and an energy optimization scheduling module;

[0009] The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the enterprise, collect energy consumption data in the enterprise in real time, preprocess the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage;

[0010] The data analysis module is used to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX according to the energy data set S;

[0011] The early warning module is used to obtain the comprehensive energy consumption score PF of the enterprise according to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, and preset the energy consumption score threshold PFYZ, compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF, and evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be in an abnormal state, and the automatic alarm mechanism is triggered to adjust the energy consumption;

[0012] The energy consumption prediction module is used to construct an energy consumption prediction model using the ARIMA algorithm when the energy consumption of the enterprise is evaluated to be in a normal state, and obtain the energy consumption E of the enterprise in the future period according to the energy consumption prediction model;

[0013] The energy optimization scheduling module is used to perform energy optimization scheduling based on the energy consumption E of the enterprise in the future, and record the energy consumption data of the enterprise after energy optimization scheduling, combine it with the energy data set S, perform summary calculation, obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ, compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ, and evaluate the effect of energy optimization scheduling.

[0014] Preferably, the data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit;

[0015] The data collection unit is used to deploy a smart sensor group in the enterprise to collect the energy consumption data of the enterprise, wherein the energy consumption data includes power consumption data, water consumption data, equipment operation data and environmental data;

[0016] The power consumption data includes the power usage of various areas and equipment in the enterprise, and is acquired by using Modbus protocol meters to collect real-time power consumption data;

[0017] The water consumption data includes the water consumption of each area within the enterprise, which is obtained by using ultrasonic flow meters and smart water meters to collect real-time water flow data;

[0018] The equipment operation data includes the switch status, operation load and working cycle of the equipment, which are acquired by installing intelligent sensors on the equipment to collect the equipment's working status data in real time;

[0019] The environmental data includes temperature and humidity data of various areas within the enterprise, which are obtained by using temperature and humidity sensors.

[0020] Preferably, the preprocessing unit is used to preprocess the collected energy consumption data, wherein the preprocessing includes data consistency check, data anomaly detection and repair, and data dimension reduction;

[0021] The data consistency check refers to aligning the time and unit of the collected energy consumption data, and converting all data into a time series structure;

[0022] The data anomaly detection and repair refers to using the Z-score method to detect outliers in energy consumption data and remove them to prevent them from affecting the analysis results;

[0023] The data dimension reduction refers to using principal component analysis technology PCA to extract core information from energy consumption data and compress multidimensional data features into principal component features.

[0024] Preferably, the data analysis module is used to perform summary calculations based on the energy data set S to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX;

[0025] The power consumption fluctuation index DL is obtained in the following manner:

[0026]

[0027] Where n represents the total number of days of data collection, i = [1, 2, 3, ..., n], R i represents the electricity consumption on the i-th day, Indicates the mean value of electricity consumption;

[0028] The water resource consumption index WR is obtained as follows:

[0029]

[0030] In the formula, WR i represents the water consumption on the i-th day, ΔWR i represents the standby water resource consumption on the i-th day, where the standby water resource consumption refers to the water resource consumption of the cooling equipment when it is not working.

[0031] Preferably, the equipment energy efficiency index NX is obtained in the following manner:

[0032]

[0033] Where P k represents the production value of the kth device, Y k represents the energy conversion efficiency of the kth device, L k represents the equipment load rate of the kth equipment, W k represents the ambient temperature of the kth device, m represents the total number of devices, k = 1, 2, 3, ..., m];

[0034] The energy conversion efficiency Y of the kth device k Obtained through the energy efficiency monitoring sensor of the equipment;

[0035] The production output value P of the kth device k Obtained through the device generation management system;

[0036] The equipment load rate L of the kth equipment k The equipment operation control system PLC is used to record the equipment's real-time operation load in real time, and then divided by the equipment's rated load to obtain the load.

[0037] Preferably, the early warning module includes a comprehensive energy consumption score calculation unit and an evaluation unit;

[0038] The comprehensive energy consumption score calculation unit is used to perform summary calculation based on the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX to obtain the comprehensive energy consumption score PF of the enterprise. The comprehensive energy consumption score PF is obtained in the following manner:

[0039] PF=α1·DL+α2·WR+α3·NX+C;

[0040] In the formula, α1, α2 and α3 represent the weight coefficients of the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX respectively, and C represents the first correction coefficient. The specific value of the weight coefficient is set by the customer according to the actual situation.

[0041] Preferably, the evaluation unit is used to preset an energy consumption score threshold PFYZ, and compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF to evaluate the energy consumption status of the enterprise. The specific evaluation contents are as follows:

[0042] If the comprehensive energy consumption score PF is less than the energy consumption score threshold PFYZ, that is, PF<PFYZ, the energy consumption of the enterprise is judged to be in a normal state, and the energy consumption data of the enterprise is recorded at this time, and the energy consumption data is stored in the enterprise energy database;

[0043] If the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, that is, PF≥PFYZ, the enterprise's energy consumption is judged to be in an abnormal state. At this time, the alarm mechanism is automatically triggered, an alarm message is generated, and the energy management personnel are notified until the energy management personnel respond. At the same time, the abnormal equipment is stopped, and the air-conditioning temperature is controlled to 70% of the original temperature. An energy consumption abnormality assessment report is generated, wherein the energy consumption abnormality assessment report includes the cause of the abnormal energy consumption status assessment and the specific time, area and equipment information of the abnormality.

[0044] Preferably, the energy consumption prediction module is used to predict the energy consumption E when the energy consumption of the enterprise is in a normal state. The specific prediction process is as follows:

[0045] According to the enterprise energy database, historical energy consumption data is collected to construct a historical energy data set H, and the historical energy data set H is divided into a training set and a validation set in chronological order, where the first 80% of the historical energy consumption data is used as the training set, and the last 20% of the historical energy consumption data is used as the validation set;

[0046] Based on the historical energy data set H, a time series graph is drawn and an ADF test is performed on the time series graph to determine the stability of the time series. If the time series is not stable, a differential operation is performed. The differential operation and ADF test are repeated until the time series is stable and the number of differentials d is obtained;

[0047] Draw the ACF and PACF diagrams of the stationary time series, and tail and truncate the ACF and PACF diagrams to obtain the order p of the autoregressive term and the order q of the moving average term;

[0048] According to the difference number d, the order of autoregressive term p and the order of moving average term q, the ARIMA algorithm is used to build an energy consumption prediction model;

[0049] The energy consumption prediction model is specifically expressed as follows:

[0050] E(t)=b+φ1·E(t-1)+φ2·E(t-2)+...+φ p ·E(tp)+θ1·εt-1 +θ2·ε t-2 +...+θ q ·ε t-q +ε t ;

[0051] Where E(t) represents the energy consumption at the predicted time point t, E(t-1), E(t-2) and E(tp) represent the energy consumption at the time points t-1, t-2 and tp respectively, [φ1, φ2, …, φ p ] represents the autoregressive coefficient, [θ1, θ2, …, θ q ] represents the moving average coefficient, ε t-1 , ε t-2 and ε t-q represents the prediction error at time points t-1, t-2 and tq, and b represents the constant term;

[0052] The above autoregressive coefficients [φ1, φ2, …, φ p ], moving average coefficient [θ1, θ2, …, θ q ], prediction error ε t And the constant term b is obtained by the maximum likelihood estimation method MLE in the process of using historical energy consumption data to carry out energy consumption prediction model;

[0053] The training set is input into the energy consumption prediction model, the energy consumption prediction model is trained by the maximum likelihood estimation method MLE, and the energy consumption prediction model is verified by the validation set to optimize the parameters of the energy consumption prediction model;

[0054] Use the verified energy consumption prediction model to obtain the energy consumption E of the enterprise in the future.

[0055] Preferably, the energy optimization scheduling module is used to obtain the energy consumption E of the enterprise in the future period of time based on the energy consumption prediction model, and generate the energy optimization scheduling plan of the enterprise based on the energy consumption E of the enterprise in the future period of time and the energy data set S, and perform energy optimization scheduling according to the energy optimization scheduling plan, wherein the energy optimization scheduling plan includes the energy allocation of each area and each equipment in the enterprise, the execution time node and scheduling measures of the energy scheduling, and collects the actual energy consumption data of the enterprise after the energy optimization scheduling in real time;

[0056] According to the actual energy consumption data of the enterprise after energy optimization and scheduling and the energy data set S, the energy consumption change rate NY is obtained, wherein the energy data set S includes the energy consumption data collected before energy scheduling optimization, and the energy consumption change rate NY is obtained as follows:

[0057]

[0058] In the formula, E forecast (g) represents the energy consumption of the enterprise at time point g before energy optimization scheduling, E opt (g) represents the energy consumption of the enterprise at time point g after energy optimization scheduling, T represents the time period for energy optimization effect evaluation, g = [1, 2, 3, ..., T];

[0059] The energy consumption change rate threshold YZ is preset, and the energy consumption change rate NY is compared with the energy consumption change rate threshold YZ to evaluate the energy scheduling optimization effect. The specific evaluation contents are as follows:

[0060] If the energy consumption change rate NY is less than the energy consumption change rate threshold YZ, that is, NY<YZ, it is judged that the energy scheduling optimization effect is unqualified. At this time, the energy scheduling optimization strategy is recorded, and professionals are arranged to re-optimize the energy scheduling and generate an energy scheduling optimization report, in which the energy scheduling optimization report includes the evaluation results of energy optimization scheduling, energy consumption data before and after energy optimization scheduling, reasons for unqualified energy scheduling optimization, and suggestions for adjusting the energy optimization scheduling strategy;

[0061] If the energy consumption change rate NY is greater than or equal to the energy consumption change rate threshold YZ, that is, NY ≥ YZ, the energy scheduling optimization effect is judged to be qualified, the energy scheduling optimization strategy is recorded, and the energy scheduling optimization strategy is stored in the enterprise's energy database.

[0062] An Internet-based intelligent analysis method for enterprise energy usage behavior includes the following steps:

[0063] Step 1: deploy an intelligent sensor group in the enterprise to collect energy consumption data in the enterprise in real time, pre-process the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage;

[0064] Step 2: According to the energy data set S, obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX;

[0065] Step 3: According to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, the comprehensive energy consumption score PF of the enterprise is obtained, and the energy consumption score threshold PFYZ is preset, and the energy consumption score threshold PFYZ is compared and analyzed with the comprehensive energy consumption score PF to evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be abnormal, and the automatic alarm mechanism is triggered to adjust the energy consumption;

[0066] Step 4: When the energy consumption of the enterprise is evaluated as normal, the ARIMA algorithm is used to build an energy consumption prediction model, and the energy consumption E of the enterprise in the future period is obtained based on the energy consumption prediction model;

[0067] Step 5. Perform energy optimization scheduling based on the enterprise's energy consumption E in the future, and record the enterprise's energy consumption data after energy optimization scheduling. Combined with the energy data set S, perform summary calculations to obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ. Compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ to evaluate the effect of energy optimization scheduling.

[0068] The present invention provides an Internet-based enterprise energy usage behavior intelligent analysis method and system, which has the following beneficial effects:

[0069] (1) The present invention deploys an intelligent sensor group in the enterprise to collect real-time data on electricity, water resources, equipment operating status and environmental parameters in the enterprise, and constructs a high-quality energy data set S through data consistency check, anomaly detection and repair, and principal component analysis (PCA). Compared with traditional manual records or single-dimensional data collection methods, the present invention can more comprehensively and accurately reflect the energy usage of the enterprise, providing a reliable data basis for subsequent intelligent analysis and optimization decisions. At the same time, the system has real-time monitoring and response capabilities for dynamic changes in energy consumption behavior, thereby improving the intelligence level of enterprise energy management.

[0070] (2) By constructing the electricity consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX, combined with the weighted comprehensive energy consumption score PF and the energy consumption score threshold PFYZ, it is possible to quickly and accurately identify abnormal conditions in energy use. The early warning module supports a multi-level early warning mechanism. When an abnormal situation occurs, it automatically triggers the alarm mechanism and generates a detailed energy anomaly report, helping enterprises to accurately locate problem areas, equipment and causes, avoiding the problems of response lag and diagnostic ambiguity in traditional energy anomaly detection methods, thereby improving the efficiency and accuracy of enterprise energy anomaly handling.

[0071] (3) An energy consumption prediction model is constructed using the ARIMA algorithm. Based on historical energy consumption data, the energy consumption E of the enterprise in the future is predicted. An accurate energy allocation and scheduling plan is formulated through the energy optimization scheduling module to allocate energy to various areas and equipment of the enterprise. The energy consumption data after optimized scheduling is further used to evaluate the energy scheduling optimization effect. Through quantitative analysis of the energy consumption change rate NY, the advantages and disadvantages of the optimization plan are identified and continuously improved. Compared with the traditional static scheduling method, the present invention can dynamically adjust the enterprise's energy use strategy, reduce energy waste, and improve energy utilization efficiency, providing a scientific basis for the enterprise's energy conservation and consumption reduction, while helping the enterprise achieve its sustainable development goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 This is a block diagram of an Internet-based enterprise energy usage behavior intelligent analysis system of the present invention.

[0073] Figure 2 The present invention is a flowchart of an Internet-based intelligent analysis method for enterprise energy usage behavior. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] Example 1

[0076] See also Figure 1 , the present invention provides an Internet-based enterprise energy usage behavior intelligent analysis system, including a data acquisition and preprocessing module, a data analysis module, an early warning module, an energy consumption prediction module, and an energy optimization scheduling module;

[0077] The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the enterprise, collect energy consumption data in the enterprise in real time, preprocess the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage;

[0078] The data analysis module is used to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX according to the energy data set S;

[0079] The early warning module is used to obtain the comprehensive energy consumption score PF of the enterprise according to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, and preset the energy consumption score threshold PFYZ, compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF, and evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be in an abnormal state, and the automatic alarm mechanism is triggered to adjust the energy consumption;

[0080] The energy consumption prediction module is used to construct an energy consumption prediction model using the ARIMA algorithm when the energy consumption of the enterprise is evaluated to be in a normal state, and obtain the energy consumption E of the enterprise in the future period according to the energy consumption prediction model;

[0081] The energy optimization scheduling module is used to perform energy optimization scheduling based on the energy consumption E of the enterprise in the future, and record the energy consumption data of the enterprise after energy optimization scheduling, combine it with the energy data set S, perform summary calculation, obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ, compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ, and evaluate the effect of energy optimization scheduling.

[0082] In the embodiment, through the data collection and preprocessing module, the intelligent sensor group is deployed in real time to comprehensively collect the power consumption, water resource use, equipment operation status and environmental parameter data in the enterprise, and build a high-quality energy data set S, which overcomes the problems of incomplete data collection and delayed update in traditional energy management. At the same time, the preprocessing module uses data consistency check, anomaly detection and repair, and principal component analysis technology to time align, unify units, and reduce dimensions of the collected data, effectively improving the accuracy and analysis efficiency of the data. The data analysis module accurately captures the volatility and efficiency of energy use through quantitative analysis of the power consumption fluctuation index DL, the water resource consumption index WR, and the equipment energy efficiency index NX. Combined with the early warning module, the energy anomaly is detected by comparing the comprehensive energy consumption score PF with the threshold PFYZ. Rapid detection of normal states and automatic triggering of alarm mechanisms provide detailed abnormality assessment reports and adjustment suggestions, improving abnormal response speed and problem location capabilities. In the energy consumption prediction module, the ARIMA algorithm is used to build an energy consumption prediction model based on historical data to obtain the energy consumption E for a period of time in the future, providing a scientific basis for the company's precise scheduling. The energy optimization scheduling module formulates an energy allocation plan based on the predicted energy consumption E, adjusts resource scheduling in real time, and optimizes energy utilization efficiency. Combined with the dynamic evaluation mechanism of the energy consumption change rate NY, the system can iteratively optimize the scheduling strategy, realizing closed-loop management from monitoring to optimization, improving the intelligence and refinement of the company's energy management, reducing energy waste, and improving resource utilization efficiency, providing comprehensive support for the company's energy conservation and consumption reduction and sustainable development.

[0083] Example 2

[0084] Please refer to Figure 1 ,Specifically: the data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit;

[0085] The data collection unit is used to deploy a smart sensor group in the enterprise to collect the energy consumption data of the enterprise, wherein the energy consumption data includes power consumption data, water consumption data, equipment operation data and environmental data;

[0086] The power consumption data includes the power usage of various areas and equipment in the enterprise, and is acquired by using Modbus protocol meters to collect real-time power consumption data;

[0087] The water consumption data includes the water consumption of each area within the enterprise, which is obtained by using ultrasonic flow meters and smart water meters to collect real-time water flow data;

[0088] The equipment operation data includes the switch status, operation load and working cycle of the equipment, which are acquired by installing intelligent sensors on the equipment to collect the equipment's working status data in real time;

[0089] The environmental data includes temperature and humidity data of various areas within the enterprise, which are obtained by using temperature and humidity sensors.

[0090] The preprocessing unit is used to preprocess the collected energy consumption data, wherein the preprocessing includes data consistency check, data anomaly detection and repair, and data dimension reduction;

[0091] The data consistency check refers to aligning the time and unit of the collected energy consumption data, and converting all data into a time series structure;

[0092] The data anomaly detection and repair refers to using the Z-score method to detect outliers in energy consumption data and remove them to prevent them from affecting the analysis results;

[0093] The data dimension reduction refers to using principal component analysis technology PCA to extract core information from energy consumption data and compress multidimensional data features into principal component features.

[0094] In the embodiment, by deploying a smart sensor group in the enterprise, multi-dimensional data of energy consumption is comprehensively collected, including power consumption data, water resource consumption data, equipment operation data and environmental data, to ensure all-round monitoring of the enterprise's energy use behavior. Among them, high-precision collection equipment such as Modbus protocol electric meters, ultrasonic flow meters, smart water meters and temperature and humidity sensors are used to record the energy use and related environmental information of each area and equipment in real time and accurately, providing high-quality data support for energy management. In addition, the pre-processing unit performs consistency check, anomaly detection and repair, and dimensionality reduction processing on the data, which improves the standardization and integrity of the data. The data consistency check solves the problem of inconsistent timestamps and units, so that information from multiple data sources can be integrated on the same time axis; data anomaly detection and repair uses the Z score method to accurately identify and process outliers to avoid abnormal data interfering with the analysis results; data dimensionality reduction extracts key features through principal component analysis (PCA), effectively reduces redundant data, and improves data processing efficiency and model calculation performance. Overall, through the data acquisition and pre-processing module, an accurate and reliable data foundation is laid for subsequent energy analysis, prediction and optimization, and the scientificity and intelligence level of enterprise energy management are improved.

[0095] Example 3

[0096] Please refer to Figure 1 ,Specifically: the data analysis module is used to perform summary calculation based on the ,energy data set S, to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX;

[0097] The power consumption fluctuation index DL is obtained in the following manner:

[0098]

[0099] Where n represents the total number of days of data collection, i = [1, 2, 3, ..., n], R i represents the electricity consumption on the i-th day, and R represents the mean electricity consumption;

[0100] A high power consumption fluctuation index DL value indicates that the power load fluctuates greatly, the system fails to effectively balance the load, and there is waste or improper energy management. A low power consumption fluctuation index DL value indicates that the power load is relatively stable and the energy utilization efficiency is high;

[0101] The water resource consumption index WR is obtained as follows:

[0102]

[0103] In the formula, WR i represents the water consumption on the i-th day, ΔWR iIt represents the standby water consumption on the i-th day. The standby water consumption refers to the water consumption of the cooling device in a non-working state. In the non-working state, the cooling device needs to maintain a minimum cooling water circulation to prevent heat accumulation inside the device.

[0104] The equipment energy efficiency index NX is obtained in the following way:

[0105]

[0106] Where P k represents the production value of the kth device, Y k represents the energy conversion efficiency of the kth device, L k represents the equipment load rate of the kth equipment, W k represents the ambient temperature of the kth device, m represents the total number of devices, k = 1, 2, 3, ..., m];

[0107] The energy conversion efficiency Y of the kth device k Obtained through the energy efficiency monitoring sensor of the equipment;

[0108] The production output value P of the kth device k Obtained through the device generation management system;

[0109] The equipment load rate L of the kth equipment k The equipment operation control system PLC is used to record the equipment's real-time operation load in real time, and then divided by the equipment's rated load to obtain the load.

[0110] In the embodiment, through the data analysis module, according to the energy data set S, the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX are calculated to realize the accurate analysis and quantitative evaluation of the energy consumption behavior of the enterprise. The power consumption fluctuation index DL analyzes the fluctuation of the average daily power consumption and evaluates the stability of power use and the load balance, which helps the enterprise to optimize the power structure and reduce the energy waste caused by the peak-valley difference. The water resource consumption index WR deducts the water resource consumption of the cooling equipment in the standby state, quantifies the effective utilization of water resources in actual work, accurately identifies the high water consumption links, and improves the water-saving management ability. The equipment energy efficiency index NX comprehensively considers the production output value, energy conversion efficiency, operation load rate and ambient temperature of the equipment, scientifically evaluates the equipment operation efficiency and energy consumption performance, helps enterprises discover inefficient equipment and optimize the operation strategy. Through these multi-dimensional energy consumption analysis indicators, the present invention can intuitively display the key problem areas and improvement directions in energy use, provide a scientific basis for energy conservation and consumption reduction of enterprises, improve the refinement level of energy management, and provide a solid quantitative foundation for subsequent abnormal warning and optimized scheduling.

[0111] Example 4

[0112] Please refer to Figure 1 ,Specifically: the early warning module includes a comprehensive energy consumption score calculation unit and an evaluation unit;

[0113] The comprehensive energy consumption score calculation unit is used to perform summary calculation based on the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX to obtain the comprehensive energy consumption score PF of the enterprise. The comprehensive energy consumption score PF is obtained in the following manner:

[0114] PF=α1·DL+α2·WR+α3·NX+C;

[0115] In the formula, α1, α2 and α3 represent the weight coefficients of the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX respectively, and C represents the first correction coefficient. The specific value of the weight coefficient is set by the customer according to the actual situation, 0<α1<1, 0<α2<1, 0<α3<1, and α1+α2+α3=1.

[0116] The evaluation unit is used to preset the energy consumption score threshold PFYZ, and compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF to evaluate the energy consumption status of the enterprise. The specific evaluation contents are as follows:

[0117] If the comprehensive energy consumption score PF is less than the energy consumption score threshold PFYZ, that is, PF<PFYZ, the energy consumption of the enterprise is judged to be in a normal state, and the energy consumption data of the enterprise is recorded at this time, and the energy consumption data is stored in the enterprise energy database;

[0118] If the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, that is, PF≥PFYZ, the enterprise's energy consumption is judged to be in an abnormal state. At this time, the alarm mechanism is automatically triggered, an alarm message is generated, and the energy management personnel are notified until the energy management personnel respond. At the same time, the abnormal equipment is stopped, and the air-conditioning temperature is controlled to 70% of the original temperature. An energy consumption abnormality assessment report is generated, wherein the energy consumption abnormality assessment report includes the cause of the abnormal energy consumption status assessment and the specific time, area and equipment information of the abnormality.

[0119] In the embodiment, the intelligent evaluation and efficient early warning of the energy consumption status of the enterprise are realized through the comprehensive energy consumption score calculation unit and the evaluation unit. The comprehensive energy consumption score calculation unit dynamically calculates the comprehensive energy consumption score PF of the enterprise based on the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX indicators, combined with the user-defined weight coefficient and correction parameter. Compared with the traditional single energy consumption indicator evaluation method, this method can more comprehensively reflect the efficiency and stability of the overall energy use of the enterprise. Through real-time comparison with the preset energy consumption score threshold PFYZ, the evaluation unit accurately determines whether the energy consumption is abnormal and triggers the automatic alarm mechanism. When an abnormal state occurs, the system can generate a detailed evaluation report containing the cause of the abnormality, specific time, abnormal area and equipment information, assisting energy management personnel to quickly locate the problem and optimize the equipment operation status, such as adjusting the air conditioning load, reducing energy waste and potential operation risks. Compared with traditional manual detection or lagging abnormal management methods, the present invention improves the accuracy, response speed and operability of energy consumption abnormality detection, and effectively helps enterprises achieve intelligent energy-saving management goals.

[0120] Example 5

[0121] Please refer to Figure 1 Specifically: the energy consumption prediction module is used to predict the energy consumption E when the energy consumption of the enterprise is in a normal state. The specific prediction process is as follows:

[0122] According to the enterprise energy database, historical energy consumption data is collected to construct a historical energy data set H, and the historical energy data set H is divided into a training set and a validation set in chronological order, where the first 80% of the historical energy consumption data is used as the training set, and the last 20% of the historical energy consumption data is used as the validation set;

[0123] Based on the historical energy data set H, a time series graph is drawn and an ADF test is performed on the time series graph to determine the stability of the time series. If the time series is not stable, a differential operation is performed. The differential operation and ADF test are repeated until the time series is stable and the number of differentials d is obtained;

[0124] The ADF test is a statistical test method used to determine whether a time series is a stationary series. If the time series is a non-stationary series, it is necessary to convert the non-stationary series into a stationary series through the difference method before further time series modeling can be performed;

[0125] Draw the ACF and PACF diagrams of the stationary time series, and tail and truncate the ACF and PACF diagrams to obtain the order p of the autoregressive term and the order q of the moving average term;

[0126] The ACF graph is a graph used to describe the autocorrelation of a time series at different lag orders, and the PACF graph describes the correlation between the current energy consumption value and the energy consumption value of a specific lag order, excluding the influence of other intermediate lag terms;

[0127] The ACF and PACF plots can help select the order p of the autoregressive term and the order q of the moving average term in the ARIMA model, and by observing the characteristics of these two plots, the correlation pattern of different lag orders in the time series can be determined;

[0128] The ARIMA model is constructed by the order p of the autoregressive term, the order q of the moving average term, and the number of differences d. The characteristics of the ARIMA model include that both the ACF graph and the PACF graph show tailing. It is necessary to determine the parameters of the final output by trying different values ​​of the order p of the autoregressive term and the order q of the moving average term and comparing the model effects.

[0129] According to the difference number d, the order of autoregressive term p and the order of moving average term q, the ARIMA algorithm is used to build an energy consumption prediction model;

[0130] The ARIMA algorithm is a statistical modeling method widely used in time series analysis and prediction. The ARIMA model can effectively capture the trend, seasonality and random fluctuation characteristics of the time series by combining autoregression AR, difference I and moving average MA. It is used to predict the energy consumption E in the future based on historical energy consumption data.

[0131] The energy consumption prediction model is specifically expressed as follows:

[0132] E(t)=b+φ1·E(t-1)+φ2·E(t-2)+...+φ p ·E(tp)+θ1·ε t-1 +θ2·ε t-2 +...+θ q ·ε t-q +ε t ;

[0133] Where E(t) represents the energy consumption at the predicted time point t, E(t-1), E(t-2) and E(tp) represent the energy consumption at the time points t-1, t-2 and tp respectively, [φ1, φ2, …, φ p ] represents the autoregressive coefficient, [θ1, θ2, …, θ q ] represents the moving average coefficient, ε t-1 , ε t-2 and ε t-qrepresents the prediction error at time points t-1, t-2 and tq, b represents the constant term, where the autoregressive coefficient φ is used to describe the linear relationship between the current energy consumption E(t) of the time series and the past energy consumption, reflecting the influence of the past energy consumption on the current energy consumption; the moving average coefficient is used to describe the relationship between the current energy consumption E(t) of the time series and the past error term, reflecting the influence of the past prediction error on the current energy consumption E(t);

[0134] The above autoregressive coefficients [φ1, φ2, …, φ p ], moving average coefficient [θ1, θ2, …, θ q ], prediction error ε t And the constant term b is obtained by the maximum likelihood estimation method MLE in the process of using historical energy consumption data to carry out energy consumption prediction model;

[0135] The training set is input into the energy consumption prediction model, the energy consumption prediction model is trained by the maximum likelihood estimation method MLE, and the energy consumption prediction model is verified by the validation set to optimize the parameters of the energy consumption prediction model;

[0136] Use the verified energy consumption prediction model to obtain the energy consumption E of the enterprise in the future.

[0137] In the embodiment, by using the ARIMA algorithm to construct an energy consumption prediction model, the energy consumption E of the enterprise in the future can be accurately predicted, providing a scientific basis for the enterprise's energy management. Compared with the traditional simple empirical prediction or single-dimensional analysis method, this module introduces a systematic method of time series analysis in the prediction process, including ADF test, difference operation, and analysis of ACF and PACF diagrams, to ensure the stability of the time series and the accuracy of the model parameters. The energy consumption prediction model parameters are optimized by the maximum likelihood estimation method MLE, which improves the prediction accuracy and generalization ability of the energy consumption prediction model. In addition, dividing the historical data into a training set and a validation set, and verifying and optimizing the model can effectively avoid the overfitting problem and ensure the reliability and applicability of the prediction results in practical applications. The prediction results provide key input data for the enterprise's energy optimization scheduling module, making energy scheduling more accurate and efficient, reducing energy waste, and improving energy utilization efficiency. At the same time, it provides strong technical support for the enterprise's energy conservation and consumption reduction goals. Through the energy consumption prediction module, the enterprise can plan resource allocation in advance, prevent abnormal energy use, and improve the intelligence level and decision-making ability of energy management.

[0138] Example 6

[0139] Please refer to Figure 1Specifically: the energy optimization scheduling module is used to obtain the energy consumption E of the enterprise in the future period according to the energy consumption prediction model, and generate the energy optimization scheduling plan of the enterprise according to the energy consumption E of the enterprise in the future period and the energy data set S, and perform energy optimization scheduling according to the energy optimization scheduling plan, wherein the energy optimization scheduling plan includes the energy allocation of each area and each equipment in the enterprise, the execution time node and scheduling measures of the energy scheduling, and collects the actual energy consumption data of the enterprise after the energy optimization scheduling in real time;

[0140] According to the actual energy consumption data of the enterprise after energy optimization and scheduling and the energy data set S, the energy consumption change rate NY is obtained, wherein the energy data set S includes the energy consumption data collected before energy scheduling optimization, and the energy consumption change rate NY is obtained as follows:

[0141]

[0142] In the formula, E forecast (g) represents the energy consumption of the enterprise at time point g before energy optimization scheduling, E opt (g) represents the energy consumption of the enterprise at time point g after energy optimization scheduling, T represents the time period for energy optimization effect evaluation, g = [1, 2, 3, ..., T];

[0143] The energy consumption change rate threshold YZ is preset, and the energy consumption change rate NY is compared with the energy consumption change rate threshold YZ to evaluate the energy scheduling optimization effect. The specific evaluation contents are as follows:

[0144] If the energy consumption change rate NY is less than the energy consumption change rate threshold YZ, that is, NY<YZ, it is judged that the energy scheduling optimization effect is unqualified. At this time, the energy scheduling optimization strategy is recorded, and professionals are arranged to re-optimize the energy scheduling and generate an energy scheduling optimization report, in which the energy scheduling optimization report includes the evaluation results of energy optimization scheduling, energy consumption data before and after energy optimization scheduling, reasons for unqualified energy scheduling optimization, and suggestions for adjusting the energy optimization scheduling strategy;

[0145] If the energy consumption change rate NY is greater than or equal to the energy consumption change rate threshold YZ, that is, NY ≥ YZ, the energy scheduling optimization effect is judged to be qualified, the energy scheduling optimization strategy is recorded, and the energy scheduling optimization strategy is stored in the enterprise's energy database.

[0146] In the embodiment, by combining the energy consumption E and energy data set S of the enterprise in the future period generated by the energy consumption prediction model, the energy allocation and scheduling plan for each area and equipment in the enterprise is accurately formulated, including specific execution time nodes and scheduling measures. During the actual execution of the optimized scheduling plan, the optimized energy consumption data is collected and recorded in real time, and dynamically compared with the data before scheduling, the energy consumption change rate NY is calculated, and the energy consumption change rate threshold YZ is set to achieve a quantitative evaluation of the scheduling optimization effect. When the optimization effect is unqualified, the system automatically generates a detailed energy scheduling optimization report, including a comparison of energy consumption before and after scheduling, analysis of unqualified reasons and improvement suggestions, to assist professionals in further adjusting the scheduling strategy; when the optimization effect is qualified, the system stores the optimization strategy in the enterprise energy database to provide experience support for future optimization. Compared with the traditional static energy allocation method, this module is based on dynamic prediction and real-time adjustment, which reduces energy waste, improves energy utilization efficiency and the economy of enterprise operation, and has the ability to continuously improve, helping enterprises to build a scientific and intelligent energy management system, and providing strong technical support for enterprises to achieve green production and energy saving and consumption reduction goals.

[0147] Example 7

[0148] Please refer to Figure 2 ,Specifically: A method for intelligent analysis of enterprise energy usage behavior based on the Internet, comprising the following steps,

[0149] Step 1: deploy an intelligent sensor group in the enterprise to collect energy consumption data in the enterprise in real time, pre-process the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage;

[0150] Step 2: According to the energy data set S, obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX;

[0151] Step 3: According to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, the comprehensive energy consumption score PF of the enterprise is obtained, and the energy consumption score threshold PFYZ is preset, and the energy consumption score threshold PFYZ is compared and analyzed with the comprehensive energy consumption score PF to evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be abnormal, and the automatic alarm mechanism is triggered to adjust the energy consumption;

[0152] Step 4: When the energy consumption of the enterprise is evaluated as normal, the ARIMA algorithm is used to build an energy consumption prediction model, and the energy consumption E of the enterprise in the future period is obtained based on the energy consumption prediction model;

[0153] Step 5. Perform energy optimization scheduling based on the enterprise's energy consumption E in the future, and record the enterprise's energy consumption data after energy optimization scheduling. Combined with the energy data set S, perform summary calculations to obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ. Compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ to evaluate the effect of energy optimization scheduling.

[0154] In the embodiment, by utilizing the intelligent sensor group deployed in the enterprise to collect power, water resources, equipment operation and environmental data in real time, and through data cleaning, anomaly detection and principal component analysis technology, a high-quality energy data set S is generated and stored in the database, providing a reliable data basis for subsequent analysis. Secondly, by obtaining the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX, the comprehensive energy consumption score PF of the enterprise is further calculated. Combined with the energy consumption score threshold PFYZ, the energy use status can be evaluated in real time and abnormal warning and automatic adjustment can be realized, which effectively improves the efficiency and accuracy of energy management. In addition, when the energy consumption evaluation is normal, the ARIMA algorithm is used to construct an energy consumption prediction model to predict the energy consumption E in the future period, and generate an energy allocation plan to optimize energy scheduling. At the same time, the optimized energy consumption data is monitored in real time, and the energy optimization scheduling effect is quantified by calculating the energy consumption change rate NY, so as to realize dynamic adjustment and continuous improvement, improve energy utilization efficiency, reduce enterprise operating costs, and provide scientific decision-making support for energy conservation and consumption reduction and sustainable development of enterprises.

[0155] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An Internet-based intelligent analysis system for enterprise energy usage behavior, characterized by: It includes data collection and preprocessing module, data analysis module, early warning module, energy consumption prediction module and energy optimization scheduling module; The data acquisition and preprocessing module is used to deploy an intelligent sensor group in the enterprise, collect energy consumption data in the enterprise in real time, preprocess the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage; The data analysis module is used to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX according to the energy data set S; The early warning module is used to obtain the comprehensive energy consumption score PF of the enterprise according to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, and preset the energy consumption score threshold PFYZ, compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF, and evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be in an abnormal state, and the automatic alarm mechanism is triggered to adjust the energy consumption; The energy consumption prediction module is used to construct an energy consumption prediction model using the ARIMA algorithm when the energy consumption of the enterprise is evaluated to be in a normal state, and obtain the energy consumption E of the enterprise in the future period according to the energy consumption prediction model; The energy optimization scheduling module is used to perform energy optimization scheduling based on the energy consumption E of the enterprise in the future, and record the energy consumption data of the enterprise after energy optimization scheduling, combine it with the energy data set S, perform summary calculation, obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ, compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ, and evaluate the effect of energy optimization scheduling.

2. According to claim 1, the Internet-based enterprise energy usage behavior intelligent analysis system is characterized by: The data acquisition and preprocessing module includes a data acquisition unit and a preprocessing unit; The data collection unit is used to deploy a smart sensor group in the enterprise to collect the energy consumption data of the enterprise, wherein the energy consumption data includes power consumption data, water consumption data, equipment operation data and environmental data; The power consumption data includes the power usage of various areas and equipment in the enterprise, and is acquired by using Modbus protocol meters to collect real-time power consumption data; The water consumption data includes the water consumption of each area within the enterprise, which is obtained by using ultrasonic flow meters and smart water meters to collect real-time water flow data; The equipment operation data includes the switch status, operation load and working cycle of the equipment, which are acquired by installing intelligent sensors on the equipment to collect the equipment's working status data in real time; The environmental data includes temperature and humidity data of various areas within the enterprise, which are obtained by using temperature and humidity sensors.

3. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 2 is characterized by: The preprocessing unit is used to preprocess the collected energy consumption data, wherein the preprocessing includes data consistency check, data anomaly detection and repair, and data dimension reduction; The data consistency check refers to aligning the time and unit of the collected energy consumption data, and converting all data into a time series structure; The data anomaly detection and repair refers to using the Z-score method to detect outliers in energy consumption data and remove them to prevent them from affecting the analysis results; The data dimension reduction refers to using principal component analysis technology PCA to extract core information from energy consumption data and compress multidimensional data features into principal component features.

4. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 3 is characterized by: The data analysis module is used to perform summary calculations based on the energy data set S to obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX; The power consumption fluctuation index DL is obtained in the following manner: Where n represents the total number of days of data collection, i = [1, 2, 3, ..., n], R i represents the electricity consumption on the i-th day, Indicates the mean value of electricity consumption; The water resource consumption index WR is obtained as follows: In the formula, WR i represents the water consumption on the i-th day, ΔWR i represents the standby water resource consumption on the i-th day, where the standby water resource consumption refers to the water resource consumption of the cooling equipment when it is not working.

5. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 4 is characterized by: The equipment energy efficiency index NX is obtained in the following way: Where P k represents the production value of the kth device, Y k represents the energy conversion efficiency of the kth device, L k represents the equipment load rate of the kth equipment, W k represents the ambient temperature of the kth device, m represents the total number of devices, k = [1, 2, 3, ..., m]; The energy conversion efficiency Y of the kth device k Obtained through the energy efficiency monitoring sensor of the equipment; The production output value P of the kth device k Obtained through the device generation management system; The equipment load rate L of the kth equipment k The equipment operation control system PLC is used to record the equipment's real-time operation load in real time, and then divided by the equipment's rated load to obtain the load.

6. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 5 is characterized by: The early warning module includes a comprehensive energy consumption score calculation unit and an evaluation unit; The comprehensive energy consumption score calculation unit is used to perform summary calculation based on the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX to obtain the comprehensive energy consumption score PF of the enterprise. The comprehensive energy consumption score PF is obtained in the following manner: PF=α1·DL+α2·WR+α3·NX+C; In the formula, α1, α2 and α3 represent the weight coefficients of the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX respectively, and C represents the first correction coefficient. The specific value of the weight coefficient is set by the customer according to the actual situation.

7. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 6 is characterized by: The evaluation unit is used to preset the energy consumption score threshold PFYZ, and compare and analyze the energy consumption score threshold PFYZ with the comprehensive energy consumption score PF to evaluate the energy consumption status of the enterprise. The specific evaluation contents are as follows: If the comprehensive energy consumption score PF is less than the energy consumption score threshold PFYZ, that is, PF<PFYZ, the energy consumption of the enterprise is judged to be in a normal state, and the energy consumption data of the enterprise is recorded at this time, and the energy consumption data is stored in the enterprise energy database; If the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, that is, PF≥PFYZ, the enterprise's energy consumption is judged to be in an abnormal state. At this time, the alarm mechanism is automatically triggered, an alarm message is generated, and the energy management personnel are notified until the energy management personnel respond. At the same time, the abnormal equipment is stopped, and the air-conditioning temperature is controlled to 70% of the original temperature. An energy consumption abnormality assessment report is generated, wherein the energy consumption abnormality assessment report includes the cause of the abnormal energy consumption status assessment and the specific time, area and equipment information of the abnormality.

8. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 7 is characterized by: The energy consumption prediction module is used to predict the energy consumption E when the energy consumption of the enterprise is in a normal state. The specific prediction process is as follows: According to the enterprise energy database, historical energy consumption data is collected to construct a historical energy data set H, and the historical energy data set H is divided into a training set and a validation set in chronological order, where the first 80% of the historical energy consumption data is used as the training set, and the last 20% of the historical energy consumption data is used as the validation set; Based on the historical energy data set H, a time series graph is drawn and an ADF test is performed on the time series graph to determine the stability of the time series. If the time series is not stable, a differential operation is performed. The differential operation and ADF test are repeated until the time series is stable and the number of differentials d is obtained; Draw the ACF and PACF diagrams of the stationary time series, and tail and truncate the ACF and PACF diagrams to obtain the order p of the autoregressive term and the order q of the moving average term; According to the difference number d, the order of autoregressive term p and the order of moving average term q, the ARIMA algorithm is used to build an energy consumption prediction model; The energy consumption prediction model is specifically expressed as follows: E(t)=b+φ1·E(t-1)+φ2·E(t-2)+...+φ p ·E(tp)+θ1·ε t-1 +θ2·e t-2 +...+θ q ·e t-q +e t ; Where E(t) represents the energy consumption at the predicted time point t, E(t-1), E(t-2) and E(tp) represent the energy consumption at the time points t-1, t-2 and tp respectively, [φ1, φ2, …, φ p ] represents the autoregressive coefficient, [θ1, θ2, …, θ q ] represents the moving average coefficient, ε t-1 , ε t-2 and ε t-q represents the prediction error at time points t-1, t-2 and tq, and b represents the constant term; The above autoregressive coefficients [φ1, φ2, …, φ p ], moving average coefficient [θ1, θ2, …, θ q ], prediction error ε t And the constant term b is obtained by the maximum likelihood estimation method MLE in the process of using historical energy consumption data to carry out energy consumption prediction model; The training set is input into the energy consumption prediction model, the energy consumption prediction model is trained by the maximum likelihood estimation method MLE, and the energy consumption prediction model is verified by the validation set to optimize the parameters of the energy consumption prediction model; Use the verified energy consumption prediction model to obtain the energy consumption E of the enterprise in the future.

9. The Internet-based enterprise energy usage behavior intelligent analysis system according to claim 8 is characterized by: The energy optimization scheduling module is used to obtain the energy consumption E of the enterprise in the future period of time based on the energy consumption prediction model, and generate the energy optimization scheduling plan of the enterprise based on the energy consumption E of the enterprise in the future period of time and the energy data set S, and perform energy optimization scheduling according to the energy optimization scheduling plan, wherein the energy optimization scheduling plan includes the energy allocation of each area and each equipment in the enterprise, the execution time node and scheduling measures of the energy scheduling, and collects the actual energy consumption data of the enterprise after the energy optimization scheduling in real time; According to the actual energy consumption data of the enterprise after energy optimization and scheduling and the energy data set S, the energy consumption change rate NY is obtained, wherein the energy data set S includes the energy consumption data collected before energy scheduling optimization, and the energy consumption change rate NY is obtained as follows: In the formula, E forecast (g) represents the energy consumption of the enterprise at time point g before energy optimization scheduling, E opt (g) represents the energy consumption of the enterprise at time point g after energy optimization scheduling, T represents the time period for energy optimization effect evaluation, g = [1, 2, 3, ..., T]; The energy consumption change rate threshold YZ is preset, and the energy consumption change rate NY is compared with the energy consumption change rate threshold YZ to evaluate the energy scheduling optimization effect. The specific evaluation contents are as follows: If the energy consumption change rate NY is less than the energy consumption change rate threshold YZ, that is, NY<YZ, it is judged that the energy scheduling optimization effect is unqualified. At this time, the energy scheduling optimization strategy is recorded, and professionals are arranged to re-optimize the energy scheduling and generate an energy scheduling optimization report, in which the energy scheduling optimization report includes the evaluation results of energy optimization scheduling, energy consumption data before and after energy optimization scheduling, reasons for unqualified energy scheduling optimization, and suggestions for adjusting the energy optimization scheduling strategy; If the energy consumption change rate NY is greater than or equal to the energy consumption change rate threshold YZ, that is, NY ≥ YZ, the energy scheduling optimization effect is judged to be qualified, the energy scheduling optimization strategy is recorded, and the energy scheduling optimization strategy is stored in the enterprise's energy database.

10. An Internet-based enterprise energy usage behavior intelligent analysis method, used to implement the Internet-based enterprise energy usage behavior intelligent analysis system as described in any one of claims 1 to 9, characterized in that: The following steps are included: Step 1: deploy an intelligent sensor group in the enterprise to collect energy consumption data in the enterprise in real time, pre-process the collected energy consumption data, construct an energy data set S, and transmit the constructed energy data set S to the enterprise's energy database for storage; Step 2: According to the energy data set S, obtain the power consumption fluctuation index DL, the water resource consumption index WR and the equipment energy efficiency index NX; Step 3: According to the power consumption fluctuation index DL, the water resource consumption coefficient WR and the equipment energy efficiency index NX, the comprehensive energy consumption score PF of the enterprise is obtained, and the energy consumption score threshold PFYZ is preset, and the energy consumption score threshold PFYZ is compared and analyzed with the comprehensive energy consumption score PF to evaluate the energy consumption of the enterprise. When the comprehensive energy consumption score PF is greater than or equal to the energy consumption score threshold PFYZ, the energy consumption of the enterprise is evaluated to be abnormal, and the automatic alarm mechanism is triggered to adjust the energy consumption; Step 4: When the energy consumption of the enterprise is evaluated as normal, the ARIMA algorithm is used to build an energy consumption prediction model, and the energy consumption E of the enterprise in the future period is obtained based on the energy consumption prediction model; Step 5. Perform energy optimization scheduling based on the enterprise's energy consumption E in the future, and record the enterprise's energy consumption data after energy optimization scheduling. Combined with the energy data set S, perform summary calculations to obtain the energy consumption change rate NY, and preset the energy consumption change rate threshold YZ. Compare and analyze the energy consumption change rate NY with the energy consumption change rate threshold YZ to evaluate the effect of energy optimization scheduling.

Images